Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00004501 IMPLEMENTATION COMPLETION AND RESULTS REPORT ON A CREDIT IN THE AMOUNT OF SDR 33 MILLION (US$50 MILLION EQUIVALENT) AND AN ADDITIONAL CREDIT IN THE AMOUNT OF SDR 6 MILLION (US$10 MILLION EQUIVALENT) TO THE REPUBLIC OF KENYA FOR A CASH TRANSFER FOR ORPHANS AND VULNERABLE CHILDREN PROJECT June 19, 2019 Social Protection and Jobs Global Practice Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (Exchange Rate Effective June 11, 2019) Currency Unit = Kenya Shillings (KSh) US$1 = KSh 101.25 US$1 = SDR 0.72 FISCAL YEAR July 1 – June 30 ABBREVIATIONS AND ACRONYMS ASAL Arid and Semi-Arid Land AF Additional Financing BOS Beneficiary Outreach Strategy BWC Beneficiary Welfare Committee CC County Coordinator CPS Country Partnership Strategy CPU Central Program Unit CCT Conditional Cash Transfer CT Cash Transfer CT-OVC Cash Transfer for Orphans and Vulnerable Children DCS Department of Children’s Services DFID Department for International Development DLI Disbursement-Linked Indicator DLR Disbursement-Linked Result DP Development Partner EFA Economic and Financial Analyses FM Financial Management G&CM Grievance and Case Management System GDP Gross Domestic Product GoK Government of Kenya HIV/AIDS Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome HSNP Hunger and Safety Net Program ICR Implementation Completion and Results Report IEC Information, Education, and Communication IOI Intermediate Outcome Indicator KIHBS Kenya Integrated Household Budget Survey KPGA Kenya Poverty and Gender Assessment KPIA Kenya Poverty and Inequality Assessment KSEIP Kenyan Social and Economic Inclusion Project M&E Monitoring and Evaluation MDTF Multi-Donor Trust Fund MGCSD Ministry of Gender, Children, and Social Development MIS Management Information System MLSP Ministry of Labor and Social Protection NEDI North and Northeastern Development Initiative NSNP National Safety Net Program NSPP National Social Protection Policy OM Operations Manual OPCT Older Person Cash Transfer OVC Orphans and Vulnerable Children PAD Project Appraisal Document PCN Project Concept Note PCK Postal Corporation of Kenya PDO Project Development Objective PforR Program for Results PIBS Program Implementation and Beneficiary Satisfaction PMT Proxy Means Test PSP Payment Service Provider PWSD-CT Persons with Severe Disabilities Cash Transfer SAU Social Assistance Unit SDSP State Department for Social Protection SIDA Swedish International Development Agency SPS Social Protection Secretariat SR Single Registry TA Technical Assistance TF Trust Fund TTL Task Team Leader UNICEF United Nations Children’s Fund VMG Vulnerable and Marginalized Group WFP World Food Program Regional Vice President: Hafez M. H. Ghanem Country Director: Carlos Felipe Jaramillo Senior Global Practice Director: Michal J. Rutkowski Acting Practice Manager: Robert S. Chase Task Team Leader: Emma S. Mistiaen ICR Main Contributor: Betty Hanan TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 6 A. CONTEXT AT APPRAISAL .........................................................................................................6 II. OUTCOME .................................................................................................................... 15 A. RELEVANCE OF PDOs ............................................................................................................15 B. ACHIEVEMENT OF PDOs (EFFICACY) ......................................................................................15 C. EFFICIENCY ...........................................................................................................................19 D. JUSTIFICATION OF OVERALL OUTCOME RATING ....................................................................21 E. OTHER OUTCOMES AND IMPACTS (IF ANY)............................................................................21 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 23 B. KEY FACTORS DURING IMPLEMENTATION .............................................................................24 IV. WORLD BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME ........................................................................................................................... 26 A. QUALITY OF MONITORING AND EVALUATION (M&E) .........................................................26 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE .....................................................28 C. WORLD BANK PERFORMANCE...............................................................................................30 D. RISK TO DEVELOPMENT OUTCOME .......................................................................................31 V. LESSONS AND RECOMMENDATIONS ............................................................................. 32 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 36 ANNEX 2. WORLD BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ............ 45 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 47 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 48 ANNEX 5. BORROWER ICR ................................................................................................... 54 ANNEX 6. SUPPORTING DOCUMENTS .................................................................................. 64 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name P111545 Kenya Cash Transfer for Orphans and Vulnerable Children Country Financing Instrument Kenya Investment Project Financing Original EA Category Revised EA Category Not Required (C) Not Required (C) Organizations Borrower Implementing Agency Republic of Kenya Ministry of Labour and Social Protection Project Development Objective (PDO) Original PDO To increase social safety net access for extremely poor OVC households and to build the capacity of the government to more effectively deliver the National Safety Net Program. Page 1 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 50,000,000 48,519,386 49,212,303 IDA-45530 79,284,446 74,979,310 74,979,310 TF-97272 10,000,000 9,951,212 9,400,064 IDA-53120 Total 139,284,446 133,449,908 133,591,677 Non-World Bank Financing 0 0 0 Borrower/Recipient 30,000,000 0 381,465,936 UK: British Department for International Development 34,000,000 0 0 (DFID) UN Children's Fund 12,000,000 0 26,400,000 Total 76,000,000 0 407,865,936 Total Project Cost 215,284,446 133,449,909 541,457,613 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 31-Mar-2009 03-Jul-2009 12-Mar-2012 31-Dec-2013 31-Dec-2018 Page 2 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 31-Oct-2013 81.12 Additional Financing Change in Implementing Agency Change in Project Development Objectives Change in Results Framework Change in Components and Cost Change in Loan Closing Date(s) Change in Safeguard Policies Triggered Change in Implementation Schedule 01-Nov-2016 122.86 Change in Implementing Agency Change in Results Framework Change in Components and Cost Change in Loan Closing Date(s) Change in Institutional Arrangements Change in Implementation Schedule 30-Oct-2018 133.67 Change in Loan Closing Date(s) KEY RATINGS Outcome Bank Performance M&E Quality Satisfactory Satisfactory Substantial RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 19-Jun-2009 Satisfactory Satisfactory 0 02 14-Dec-2009 Moderately Satisfactory Moderately Satisfactory 1.84 03 12-May-2010 Moderately Satisfactory Moderately Satisfactory 6.20 04 26-Jan-2011 Moderately Satisfactory Moderately Satisfactory 16.25 05 10-Jul-2011 Moderately Satisfactory Moderately Satisfactory 20.45 06 27-Dec-2011 Satisfactory Satisfactory 35.11 07 25-Jun-2012 Satisfactory Satisfactory 45.39 Page 3 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) 08 26-Dec-2012 Satisfactory Satisfactory 54.10 09 26-Jun-2013 Satisfactory Satisfactory 72.85 10 17-Jan-2014 Satisfactory Moderately Satisfactory 81.12 11 07-Jul-2014 Satisfactory Moderately Satisfactory 87.01 12 10-Feb-2015 Satisfactory Moderately Satisfactory 98.19 13 27-May-2015 Satisfactory Moderately Satisfactory 98.19 14 08-Dec-2015 Satisfactory Moderately Satisfactory 108.00 15 08-Apr-2016 Moderately Satisfactory Moderately Satisfactory 118.68 16 17-Nov-2016 Satisfactory Satisfactory 122.86 17 21-Jun-2017 Satisfactory Satisfactory 126.45 18 18-Jan-2018 Satisfactory Satisfactory 130.30 19 21-Sep-2018 Satisfactory Satisfactory 133.67 20 28-Dec-2018 Moderately Satisfactory Satisfactory 133.67 SECTORS AND THEMES Sectors Major Sector/Sector (%) Social Protection 100 Social Protection 80 Public Administration - Social Protection 20 Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Public Sector Management 0 Public Administration 5 Administrative and Civil Service Reform 5 Page 4 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Social Development and Protection 0 Social Protection 95 Social Safety Nets 87 Social protection delivery systems 8 ADM STAFF Role At Approval At ICR Regional Vice President: Obiageli Katryn Ezekwesili Hafez M. H. Ghanem Country Director: Johannes C.M. Zutt Carlos Felipe Jaramillo Senior Global Practice Director: Yaw Ansu Michal J. Rutkowski Practice Manager: Christopher J. Thomas Robert S. Chase Task Team Leader(s): Michael Mills Emma S. Mistiaen ICR Contributing Author: Betty Hanan Page 5 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Country Context 1. At the time of project preparation, poverty and inequality were major challenges for Kenya, and Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome (HIV/AIDS) contributed to driving households into poverty. According to the Kenya Poverty and Inequality Assessment 2009 1, almost half of Kenya’s population (around 35.5 million people) lived in poverty, and nearly 20 percent lived in extreme poverty. Many households were driven into poverty by shocks, including food price inflation, droughts, and floods, and one particular ‘shock’ was illness and death caused by HIV/AIDS. The preliminary report of the 2008 Kenya AIDS Indicator Survey showed that more than 1.4 million Kenyans were living with HIV/AIDS, with an estimated prevalence rate of 7.8 percent. Mostly due to HIV/AIDS, mortality in the most socially and economically productive age cohort (15–49 years) had increased by about 30 percent since 1998, and life expectancy at birth had fallen to about 47 years (from 58 years in 1990). With the support of the World Bank, the Government of Kenya (GoK) was strengthening its efforts to prevent the spread of HIV, but further steps had to be taken to mitigate the effects of the epidemic. 2. Governance remained a concern in Kenya. Corruption was cited as the main constraint to attracting private investment, and many Kenyans believed that poor governance and corruption contributed to Kenya’s continuing poverty. Kenya had a historically free and vibrant press and an active and engaged civil society that helped expose poor governance and demanded action to address it. Parliament had also actively engaged in keeping the GoK accountable and the GoK acted to address these issues. Measures included enhancing GoK-employee accountability through (a) placing employees on performance contracts; (b) deepening fiduciary reforms; (c) promoting transparency and accountability; and (d) preventing, investigating, and prosecuting corruption and economic crimes involving officials. At the time, the GoK was also assessing its Governance Strategy for Building a Prosperous Kenya and preparing the second phase of its Governance Action Plan. A plan for governance strengthening and anticorruption in the health sector was being implemented with assistance from development partners (DPs). It was recognized, however, that more needed to be done. Orphans and Vulnerable Children (OVC) 3. Kenya had a growing number of OVC and they were in need of support. A rapid assessment undertaken in 2004 estimated that 1.8 million children were orphans, having lost one or both parents and nearly two-thirds had lost their parents due to HIV/AIDS. The proportion was increasing with about 11 percent of all children in Kenya in 2003, compared to about 9 percent five years earlier. The National AIDS Control Council also estimated that there were about 600,000 children who were not orphans but particularly vulnerable, because their “safety, wellbeing, and development were, for a variety of reasons, threatened.” Many of these children had parents who were affected by AIDS and were unable to work or support their children. Families missing one or both parents were also mostly poorer than the general 1World Bank Group. 2009. Kenya Poverty and Inequality Assessment Executive Summary and Synthesis Report. Report No. 44190-KE. Page 6 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) population; nearly one-quarter of OVC lived in extreme poverty. Basic services to orphans was often negatively affected. For example, only 85 percent of orphans attended school, compared to 93 percent of other children. Orphans tended to start school at a later stage and drop out earlier than other children. In addition, the majority of orphans did not have birth registration documents and a survey carried out in 2004 showed that only about 15 percent of orphans were receiving some kind of health care. 4. The importance of supporting OVC was a growing concern in the country. With OVC representing almost 30 percent of all children living in poverty, and with the increasing magnitude of the problem, there was a growing recognition that these children needed additional support to prevent an “underclass” from developing, and to fully harness Kenya’s human development capacity. The burden of looking after orphans mostly resided with grandparents (40 percent) and other relatives (34 percent), with two-thirds of orphans living in female-headed households 2. Moreover, the traditional community-based response mechanisms were also being overstrained by the HIV/AIDS pandemic, by economic and political shocks (including the post-election violence), and by social change. Although families continued to provide care for OVC, the capacity of these coping mechanisms needed to be increased. 5. The GoK recognized the importance of improving its social protection (SP) polices, including increased support to OVC and several SP interventions were already in place. The GoK had finalized its Medium-Term Plan 2008–12 for its Vision 2030 and the (then) Ministry of Gender, Children, and Social Development (MGCSD) was developing an SP Policy and a National SP Strategy. In the short and medium term, the objective was to meet the immediate needs of the poorest and most vulnerable, focusing primarily on the population living in extreme poverty and those most vulnerable to income poverty due to external shocks. In the long term, the strategy aimed to facilitate the development of a comprehensive SP system. A draft national policy on OVC had also been developed aiming to ensure that all OVC were protected and supported to achieve their full potential. Based on this policy and the results from a rapid assessment of OVC, a National Plan of Action on OVC had been developed and there was a growing recognition that these children needed special attention and support. A number of SP interventions were in place, including the school feeding program and emergency food aid programs supported by WFP; a pilot Hunger Safety Net Program (HSNP) supported by the U.K. Department for International Development (DFID), providing regular and predictable amounts of cash to chronically food-insecure households in the arid and semi-arid land (ASAL) areas; a Drought Contingency Fund, supported by the World Bank and the European Commission (Arid Land Resource Management Project) and a number of other DP-supported SP initiatives focusing on OVC (DFID, United Nations Children’s Fund [UNICEF], and the World Bank). The Cash Transfer for Orphans and Vulnerable Children (CT-OVC) Program 6. During project preparation, the CT-OVC Program had been under way for four years, first with a pre-pilot and then a pilot program. The pre-pilot began in December 2004 and was supported by UNICEF and the Swedish International Development Agency (SIDA). The pre-pilot provided cash transfers (CT) to 500 poor households caring for OVC in three districts and generated lessons on program targeting, selection procedures, and implementation costs to guide the design of a pilot project, which was launched in 2006 and which expanded the program in 40 additional districts. 2About 7 percent of orphans were in foster care or were adopted, and the rest had to find ways to survive by themselves, often through living rough on the streets. Page 7 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) 7. The program had well-established operating principles. Substantial attention was given to results and efficiency, accountability, and transparency. Clearly defined criteria had been established for targeting districts, locations, and households. A baseline had been established in the pilot districts to enable an impact evaluation, which was to ensure that the CTs were reaching the most vulnerable children and assess the impact on their welfare. In addition, a communication strategy was developed to promote transparency. 8. In 2008, the program was being further expanded and the GoK was committed to strengthen program delivery with support from the World Bank through this Project. By July 2008, there were about 25,000 beneficiary households and by June 2009, the program was expected to deliver CTs to about 65,000 OVC households in 47 districts. It was acknowledged that further strengthening of the GoK’s capacity and delivery systems was required. The GoK’s aim was to reach about half of all OVC living in extreme poverty, through a nationwide program covering at least 100,000 households by 2012. The overall strategy was to intensify and enhance activities in the 40 districts that had already been included in the expanded program, as well as to ensure its continued implementation in the pilot districts. Theory of Change (Results Chain) 9. The theory of change is summarized in Table 1. 10. Under Component 1 – Policy Development and Institutional Strengthening, the theory of change was that by providing (a) more accessible and predictable CTs, (b) harmonized programs and improved sector coherence, (c) strengthened governance and program systems, and (d) strengthened capacity for program management, the Project enabled increased capacity of the GoK to more effectively deliver the National Safety Net Program (NSNP). This would be possible through, among others, the development of a policy framework to better coordinate existing SP programs and support the overall planning and coordination of SP interventions. Investments in management would ensure that the Project’s investments were well implemented, monitored, and evaluated, so that the benefits would be sustained and expanded and that future investments could benefit from lessons of experience. 11. Under Component 2 – Program Implementation in Selected sub-counties, the theory of change was that by (a) delivering benefits to targeted extremely poor households, (b) providing improved human capital outcomes for OVC households, and (c) implementing the Program effectively in selected counties, the Project would enable increased social safety net access for extremely poor OVC households. CTs would provide funding to caretakers of OVC, promoting school enrolment and attendance, regular health checkups, vaccination, and birth registration. CTs would also be a way to raise awareness of the importance of nutrition, immunization, health checkups in general, and infectious diseases including HIV/AIDS. Together, these investments would increase levels of education, vaccination and birth registration leading to increased overall child welfare in the short run and human capital and well-being in the long run. Page 8 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Table 1. Theory of Change Intermediate Activities Outputs/Intermediate Outcomes Final Outcome Outcomes • National Social Protection Strategy and Policy Framework established Harmonized • Policy framework development • NSNP Consolidation strategy developed and adopted programs and • Preparation of consolidation • NSNP Consolidation strategy implemented, including creation improved sector strategy for NSNP of Social Assistance Unit (SAU) and Harmonized Operations coherence Manual • Expansion of biometric • Payment mechanism strengthened—Increased percentage of CTs more enrolment for payment NSNP beneficiaries receiving electronic payments using two- accessible and PDO: To authentication factor authentication predictable increase social • Support for FM and safety net procurement; comprehensive access for FM manual for payments extremely poor developed OVC households • Single registry (SR) • Procurement and fiduciary systems strengthened and to build the development, FM module, and • Recruitment and deployment of additional fiduciary staff capacity of the improvement in targeting • SR operational and linked with IPRS Strengthened GoK to more mechanism • FM module in MIS linked to payment service providers governance and effectively • Harmonized targeting tool • Improved targeting mechanism for CT-OVC Project program systems deliver the developed • Harmonized targeting for the NSNP - Increased percentage of Strengthened National Safety • Impact Evaluation, Program NSNP beneficiaries who conform to program targeting criteria capacity for Net Program Implementation and • Strengthened G&CM system—G&CM system established at program Beneficiary Surveys, and national level and decentralization under way management Operational Reviews • Percentage of overall beneficiary satisfaction • Development and • Regular M&E report for the Program implementation of Beneficiary Outreach Strategy (BOS) • Headquarter and field officials trained on implementation, Page 9 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) coordination, and monitoring and evaluation (M&E) • Increased number of CT-OVC beneficiary households in 47 counties receiving cash transfers; increased overall NSNP • Identification and provision of coverage Delivery of targeted CT-OVC beneficiaries • Increased proportion of OVC in beneficiary households ages benefits to • Awareness creation for CT-OVC 6–17 years have completed or are enrolled in basic education targeted households to comply with co- • Increased proportion of OVCs living in beneficiary households extremely poor responsibilities related to 0–17 years of age holding birth certificates OVC households health and education • Beneficiary awareness increased; increased percentage of Improved human outcomes NSNP beneficiary households who know program objectives capital outcomes • Pilot of penalties for and entitlements and increased percentage who can name for OVC noncompliance two means of making a complaint • Increased proportion of OVCs in beneficiary households ages 0–5 fully immunized • Provision of operating costs related to program Program • Staff recruited to support effective program implementation administration and delivery, implemented at national and county levels including, among others, effectively in • Increased percentage of CT-OVC payments disbursed to contracting payment service selected payment service providers on time provider (PSP) for delivery of districts CTs Page 10 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Project Development Objectives (PDOs) 12. The original PDO was to increase social safety net access for extremely poor OVC households, through an effective and efficient expansion of the CT-OVC Program. 3 13. The wording of the PDO was the same in the main text of the Project Appraisal Document (PAD), annex 3 of the PAD, and in the Financing Agreement. Key Expected Outcomes and Outcome Indicators 14. The achievement of the PDO was to be measured by the following key performance indicators: • Additional number of OVC beneficiary households receiving cash transfers • Proportion of beneficiary households that met the targeting criteria • Proportion of OVC in beneficiary households aged 6–17 years who have completed or are enrolled in basic education Components 15. The Project aimed to support a scale-up of the GoK’s Program over four years (FY2010–13)4. It had two components. 16. Component 1: Policy Development and Institutional Strengthening (US$10 million or 20 percent of total project costs) which consisted of the subcomponents listed below. 17. Sub-component 1.1: Strengthening the Capacity of the MGCSD to coordinate SP interventions (US$1 million; 2 percent of total project costs). This sub-component supported the development of a policy framework to better coordinate existing SP programs and support the overall planning and coordination of SP interventions, mainly through technical assistance (TA) and training of key staff. 18. Sub-component 1.2: Strengthening the Capacity of the Department of Children Services (DCS) to more Efficiently Manage the Program (US$4 million equivalent; 8 percent of total project costs). This sub-component provided support for operational activities; TA for procurement, financial management (FM), and an update of the Operations Manual (OM); training materials; and hardware. It also supported links with complementary programs to: (a) strengthen the supply side of the program, such as links with schools and health centers, and also coordination with other human development promotion programs such as school bursary schemes, HIV/AIDS awareness programs, and health fee waiver programs; (b) 3 The system generated datasheet of this report, on page 1, is erroneously showing the revised PDO under the “original PDO”. The revised PDO is “To increase social safety net access for extremely poor OVC households and to build the capacity of the government to more effectively deliver the National Safety Net Program.” 4 The Project was designed as a CT Project but had some initial trial elements of a CCT (Para 94). PDIs selected were relevant for both CT and CCT projects. Increased education enrolment and attendance, health and vaccination are all indicators that translate into efficiency of its implementation. Page 11 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) promote income-generating activities and skills development of OVC caregivers; and (c) promote the skills development of youth, by linking with youth polytechnics and support programs. 19. Sub-component 1.3: Governance and Accountability (US$2 million; 4 percent of total project costs). This subcomponent funded: (a) provision of information to beneficiary households and the population as a whole regarding the program and its benefits; (b) implementation of a communication strategy, including the design of the program website; (c) information, communication, and education campaigns; (d) program leaflets and posters and other promotional information and materials; (e) implementation of awareness on immunization, nutrition, infectious diseases including HIV/AIDS, and other knowledge to improve the life of OVC households; and (f) enhanced oversight and accountability systems to improve program implementation and to assure that beneficiaries as well as non-beneficiaries had access to the appeals and complaints mechanism. To ensure beneficiary satisfaction, activities such as spot checks and citizen’s report cards were also financed. 20. Sub-component 1.4: Management Information, Monitoring and Evaluation (US$3 million; 6 percent of total project costs). The sub-component financed: (a) refinements to the MIS; (b) an FM module linking with payment providers; (c) technical support for operational and impact evaluation and financial strengthening processes and (d) studies and a pilot to test alternative payment mechanisms. 21. Component 2: Program Implementation in Selected Districts (US$40 million; 80 percent of total project costs). It had two sub-components, as listed below. 22. Sub-component 2.1: Cash Benefits for OVC Households in Selected Districts (US$39 million, 78 percent of total project costs). IDA funds were to cover CTs of K Sh 1,500 per month per household in 25 of the 40 GoK-funded districts and up to 56,000 households. The program was to be expanded in seven districts in the first year, reaching an estimated 20,000 households. In subsequent years, the Project was to support expansion in an additional six districts per year. In the second year, about 12,000 additional households were to be supported, followed by another 12,000 in the third year, and another 12,000 in the fourth year, aiming to reach 56,000 households by the end of FY12/13 in 25 districts. 23. Sub-component 2.2: Transaction Costs for Cash Transfer (US$1 million, 2 percent of total project cost). This sub-component financed operating costs related to program administration and delivery, including, among others, the transaction costs for the transfer of the cash from the GoK to the beneficiary households. B. SIGNIFICANT CHANGES DURING IMPLEMENTATION 24. The Project had three restructurings. The first one was in 2013 when an Additional Financing (AF) from IDA (of US$10 million) was processed, together with additional funding from DFID (of around US$ 56.4 million), expanding support from the CT-OVC program to all CT programs under the NSNP aligned with a parallel Program for Results (PforR) operation supporting the NSNP (see paragraph 30). 5 The second 5The NSNP PforR aimed at contributing to build the capacity of the GoK to deliver the NSNP. The PforR (of US$ 300 million IDA credit) consists of eleven disbursement linked indicators (DLIs) and the World Bank disburses funds upon achievement of these results. Page 12 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) one was in 2016 when the closing date was extended and other minor amendments done. The third one was in 2018, when the closing date was extended again. Details on the changes are provided in table 2. Table 2. Summary of Project Changes Restructuring Date Description IDA AF (Cr. 5312-KE) and Trust Fund TF097272. The AF (IDA) and the additional funding (DFID) was to provide, among others, TA to support achievement of the DLIs under the PforR and additional funding for CTs to beneficiaries. It (a) extended the closing date of the original project by three AF (IDA), years to December 31, 2016 and (b) made changes to the original project to Additional October reflect the broader focus of the AF (IDA) and the additional funding (DFID) to 1 Funding (DFID) 31, 2013 be aligned with the objective of the NSNP; (c) expanded the coverage of the and/ CT-OVC Program and expanded support to build GoK capacity to deliver its Restructuring NSNP; (d) modified the PDO and the description of Component 1; (e) reflected the additional funding to the TF097272; (f) adjusted the Results Framework to better reflect the focus of revised Component 1; and (g) aligned the Project with the PforR. TF097272. This restructuring: (a) extended the closing date by almost two years to October 31, 2018; (b) adjusted the Results Framework (including Novemb revision of some of the indicators linked to DLIs under the PforR to ensure 2 Restructuring er 1, alignment) to better measure key indicators and revised some of the targets; 2016 and (c) amended the institutional arrangements based on changes in the MLSP. TF097272. This restructuring extended the closing date to December 31, October 2018, to: (a) provide sufficient time for the GoK to complete outstanding 3 Restructuring 30, 2018 procurement activities and (b) conclude the analysis of PIBS2 and the CT-OVC Impact Evaluation. Revised PDOs and Outcome Targets 25. The PDO was revised to reflect the broader aims of the AF. The revised PDO was to increase social safety net access for extremely poor OVC households and build the capacity of the GoK to more effectively deliver its National Safety Net Program. 26. The wording of the revised PDO was the same in the main text of the Project Paper for the AF and its Financing Agreement. Revised PDO and Intermediate Indicators 27. The Results Framework was adjusted to better reflect the revised PDO (annex 1). As part of the restructuring in 2013, the intermediate output indicator (IOI) measuring ‘Proportion of beneficiary households reporting satisfactory Program implementation in independent external assessments’ was dropped because it proved difficult to measure and was replaced with two IOIs that were introduced to measure awareness of beneficiaries (IOI3 and IOI4). An indicator was also added to measure payments made electronically using two-factor authentication (IOI2). Other minor amendments were made to the wording and targets of some of the indicators to be aligned with the NSNP PforR. In 2016, the first PDO Page 13 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) indicator was also split into two indicators: one measuring additional beneficiaries (PDO1) and one measuring timeliness (PDO2). The targets were again amended and updated. Revised Components 28. As noted in Table 2, the restructuring in 2013, included revisions to the PDO. It supported the two original components with some modifications. 29. Component 1: Policy Development and Institutional Strengthening (US$10 million, 100 percent of IDA AF and US$5.2 million through DFID co-financing). The revised component was aligned with the objective and results under the NSNP PforR and aimed to contribute to building the capacity of the GoK to deliver the NSNP by: (a) strengthening program systems to ensure good governance; (b) harmonizing cash transfer programs to improve sector coherence; and (c) providing capacity building for expanding CT programs to promote more comprehensive and equitable coverage. The component financed: (a) TA; (b) training; (c) vehicles; (d) office and information and communication technology equipment; (e) printing of documents and materials; (f) website development and maintenance; (g) rollout of the communication strategy; and (h) operating costs including additional temporary staff. 30. Component 2: Program Implementation in Selected sub-counties (US$51.2 million equivalent through DFID co-financing). The activities that had been financed remained unchanged from the original project as they consisted of financing CT-OVC CTs. The KSh 1,500 per household per month under the original project, was increased to KSh 2,000 during implementation of the AF. Other Changes 31. The AF provided support for a range of activities under Component 1. While Component 1 originally aimed to strengthen the GoK’s capacity to develop its SP Policy, coordinate SP interventions, and manage the CT-OVC Program, the AF expanded its scope to build the GoK’s capacity to support the entire NSNP. 6 This was necessary because the GoK established the NSNP and planned to increase the financing and coverage of all of the NSNP CT programs, 7 including doubling the financing and coverage of the CT-OVC Program. The NSNP required further investments in institutional and system strengthening and support to the GoK to ensure quality of the results achieved under the NSNP PforR. The AF closing date was also established as December 31, 2016. 32. The institutional arrangements of the Project were modified. This change reflected the ministerial restructuring that reduced the number of ministries, among others. As a result, the Department of Children’s Services (DCS), which implemented the CT-OVC Program was moved from the MGCSD (which was eliminated) to the Ministry of Labor, Social Security, and Services. At the same time, local-level structures were renamed to reflect the establishment of 47 counties. Despite this shift in 6 The NSNP, supported by the World Bank, through the PforR, seeks to support Kenya to establish an effective NSNP for poor and vulnerable households through the four main CT programs in the country: the CT-OVC program, the Older Persons Cash Transfer (OPCT), the Persons with Severe Disability Cash Transfer, and the HSNP. 7 The NSNP now reaches around 765,000 households (over 4 million people), exceeding the originally expected coverage of 530,000 households by 2017. In addition, the GoK is expanding to cover 500,000 additional individuals under the enhanced OPCT, which has shifted from targeting poor households with an elderly person above 65 years of age to targeting all individuals over 70 years. As such, the total number of individuals reached through the NSNP is around 4.5 million people. Page 14 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) ministries and the name of local-level bodies, the implementation arrangements remained largely the same. In July 2016, the SAU was established within the Ministry of Labor and Social Protection (MLSP) to consolidate the management of the three CT programs. Staff were deployed to work full time in the unit. The SAU operates at the same level as other departments in the ministry. Currently, the CT-OVC Program implementation is coordinated by the SAU at the national level while implementation at county and sub- county levels is still being implemented by the Children Officers under the DCS. The DCS also has liaison officers acting as the link between the SAU and DCS. With the AF, it was also agreed that the Social Protection Secretariat (SPS) should play a more prominent role in the implementation of Component 1. Rationale for Changes and Their Implication on the Original Theory of Change 33. Changes to the PDO under the IDA AF were justified. Although the effectiveness and efficiency part of the original PDO was not removed, the revised PDO was made more ambitious. The scope was broadened from only focusing on the effectiveness and efficiency of the CT-OVC Project to focusing more broadly on Government’s capacity to effectively implement the NSNP as a whole. Thus, the AF enhanced the impact of the CT-OVC Program and the broader NSNP, which had proven to be effective in supporting poor and vulnerable households taking care of OVC. As described in paragraph 26, the PDO was modified to better reflect the aim of the AF. The Project, however continued to only finance CTs for CT-OVC beneficiaries (under the trust fund). II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating 34. The relevance of the PDOs was High. The PDO is as relevant today as it was when the original project and the AF were designed. The PDO was consistent with the GoK’s priorities as articulated in its draft SP Policy and its National SP Strategy. The strategy objectives were to harmonize SP interventions across the country while: (a) meeting the immediate needs of the poorest and most vulnerable, and (b) focusing primarily on the population living in extreme poverty and those most vulnerable to income poverty due to external shocks. In the long-term, the strategy aimed to facilitate the development of a comprehensive SP system. Similarly, the relevance of the PDO was consistent with the World Bank’s Country Assistance Strategy for Kenya 2004–08, which emphasized investing in people, reducing vulnerability, and strengthening communities. The CT-OVC Project remains fully aligned with the Country Partnership Strategy (CPS) for Kenya (2014-2020), as well as Government strategies. The CPS prioritizes “promoting protection and potential to ensure all groups share in advancing prosperity and helping the vulnerable to develop their potential”. Kenya’s Vision 2030 seeks to build “a just and cohesive society with social equity”. B. ACHIEVEMENT OF PDOs (EFFICACY) Assessment of Achievement of Each Objective/Outcome 35. The Project was successful in almost fully achieving its objectives and is rated Substantial for efficacy. Since the PDO was made more ambitious and expanded the scope to support the NSNP as a whole (of which CT-OVC program forms part) and aligned with the Implementation Completion and Page 15 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Results (ICR) guidelines, no split rating was used for this evaluation. Based on this ICR evaluation, most of the objectives/outcomes demonstrate achievement to warrant a Substantial rating. The ICR evaluation has been based on analyses of outcome indicators and data linked to the PDO, as well as an evaluation of how Project activities were linked to achievement of the PDO both under the original Project and the AF. The implementation of the Project was monitored through three PDO-level indicators and eight IOIs. Two out of the three original PDO indicators were either overachieved or achieved and one mostly achieved. Out of the three revised PDO indicators, one was achieved and two were mostly achieved. 36. In terms of relevance related to the first part of the PDO ‘increase social safety net access for extremely poor OVC households’, the Project was instrumental in helping extremely poor OVC beneficiary households through increased coverage, as demonstrated by the number of households that are being supported by the program. The GoK expanded the program well above what was initially envisioned under the original Project (original target was 56,000 households and final achieved target was 340,000 households). The original target (PDO1) was revised upwards to 365,000 as part of restructuring in December 2016. The revision was motivated by an ongoing expansion on the number of beneficiaries and the target of 365,000 was expected to be easily achieved. Although the program reached more than six times the original target, the revised target was not fully achieved before Project closure as some households had exited from the Program, and no resources had been allocated by the GoK to replace these households or for further expansion. Payments for some households were also put on hold due to pending updates of beneficiaries’ information and resolution of other G&CM cases. 37. In addition, with the rollout of new payment mechanism, CTs are more secure, accessible, and predictable in rural areas where the payment providers have engaged several agents to pay beneficiaries, thereby considerably decreasing the time and costs associated with the beneficiaries having to travel long distances and wait in long queues to receive the transfers. According to PIBS1 39 percent had to travel less than 6 km to the nearest pay point, compared to 84 percent in PIBS2. In addition, as shown in Table 3, the targets related to proportion of OVC: (i) (6-17 years of age) who have completed or are enrolled in basic education; (ii) (0-17 years of age) holding birth certificates; and (iii) (0-5 years) fully immunized have been achieved and, in two cases, overachieved. Table 3. Achievement of Targets for Part 1 of the PDO PDO Results Indicators Target Comment Objective Achieved Objective PDO1. Additional number of OVC Original: 56,000 340,000 Mostly achieved 1: To beneficiary households receiving cash Revised: increase transfers 365,000 social PDO4. Proportion of OVC in beneficiary 95 95 Achieved safety net households aged 6–17 years who have access for completed or are enrolled in basic extremely educational poor OVC IOI6. Proportion of OVCs living in 60 67 Overachieved households beneficiary households 0–17 years of age holding birth certificates IOI7. Proportion of OVC in beneficiary 70 81 Overachieved households age 0–5 years fully immunized Page 16 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Note: (a). The original baseline estimated for this indicator was 77 percent (estimated using Kenya Integrated Household Budget Survey (KIHBS) 2005/06 data for all OVC households in 25 districts). The baseline in 2012 however confirmed a value of 94 percent, but the end target remained 95 percent since it was already so high. (b). The end-line survey reported on an indicator of ‘vaccination card up to date’ which applied to 81 percent of beneficiaries and is interpreted to be the same as fully immunized for a beneficiary’s particular age. 38. Related to the second part of the PDO to ‘build the capacity of the GoK to more effectively deliver its National Safety Net Program’, the Project played a central role in strengthening capacities of GoK institutions and systems to deliver CTs effectively and efficiently to the neediest households (paragraph 38). In general, the GoK’s capacity to deliver the NSNP has greatly improved. More specifically, the introduction of an electronic payment mechanism with two-factor authentication ensured that CTs became more secure, accessible, and predictable. Further, the Project has been instrumental in strengthening the integrity of management systems by, among others, adding biometric enrolment and identification of beneficiaries, improving targeting mechanisms, and consolidating FM processes. The operation has also supported the establishment of a G&CM system to enable beneficiaries and others to lodge grievances or updates on household data. These initiatives and results addressed in paragraph 44, have helped improve the delivery of CTs to beneficiary households. 39. As shown in Table 4 below, out of eight targets, five were either mostly achieved/achieved or overachieved and three were not achieved (but two of those related to timeliness of payments). It is, however important to note that the three indicators which were not achieved, still represented significant achievements under the Project. Related to timeliness of payments, it was overall significantly improved for the Project, despite not reaching the set target. Between 2014 and 2017, NSNP payments were mostly made on time (or the delays significantly decreased, with only a few days delayed, compared to previous 7–8 months delayed) and in 2017 the GoK achieved a DLI under the PforR related to timeliness of payments. 8 The PIBS2 report (2017/18) also confirmed that 77 percent of NSNP (including OVC) beneficiaries reported having received payments on time. In 2018, the NSNP (including OVC) again suffered payment delays (and during the last three payment cycles under the Project, only 13 percent of OVC payments were made on time), but these delays were due to two main reasons: (a) the launch of the universal OPCT, which was given top priority, and payments to other NSNP programs under the SAU were put on hold to synchronize payment with the newly registered OPCT beneficiaries and (b) rollout of the new unique payment mechanism, which has now ensured timely payments for the last three payment cycles of FY2018/2019, and beneficiaries now receive a higher-quality service, because they have access to a bank account and a choice in payment provider. Related to the third target which was not achieved, beneficiaries who can name two means of making a complaint, a significant increase was still achieved with awareness raised from 8 percent to 30 percent. Although the target was not achieved for the Project, it was achieved for the NSNP PforR (for which the target was revised downwards following conclusion of a significantly lower baseline than expected - see Note (b) of Table 4). Unfortunately, due to an oversight, the target was not revised downward as it was under the PforR. 8 This DLI measured timeliness for all NSNP payment for three consecutive payment cycles. Page 17 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Table 4. Achievement of Targets for Part 2 of the PDO PDO Results Indicators Target Comment Objective Achieved Objective PDO2. Percent of all NSNP 70 13 Not achieved 2: To build payments disbursed to payment the service providers on time capacity of PDO3. Percent of NSNP 85 82 Mostly achieved the GoK to beneficiaries who conform to the more Program targeting criteriaa effectively IOI1. Social Protection Strategy SP New SP Partially achieved. An SP Policy deliver the and policy framework established Strategy Policy Framework was adopted in June National adopted Framework 2011. The SPS is currently updating Safety Net and the framework and is preparing an Program Strategy in SP strategy. A workshop on the process draft SP Policy took place in September 2018. IOI2. Percent of NSNP 90 100 Overachieved beneficiaries for whom payments are made electronically through payment services providers using two-factor authentication. IOI3. Percent of NSNP 47 30.4 Not achieved. Achieved as per the beneficiaries who can name two formally revised PforR target means of making a complaintb (30%), but the Project’s target (47%) was not revised because no restructuring was made just before the closing date. IOI4. Percent of NSNP 60 89.4 Overachieved beneficiaries who know program objectives and entitlementsc IOI5. Percent of NSNP progress 100 100 Achieved. Biannual and annual reports produced by the MIS, reports have been produced by the according to agreed reporting MIS. standards IOI8. Percent of OVC payments 60 13 Not achieved. During the last three disbursed to payment service payment cycles, only 13 percent of providers on time OVC payments were made on time. Note: (a). This indicator was dropped as part of the AF for the PforR due to complications in calculating targeting and difficulties in interpreting improvements in conforming to targeting at least related to poverty. Instead it was changed to establishing and implementing a recertification process. Nonetheless in PIBS2, it is estimated that 82.4 percent of beneficiaries conform to the targeting criteria. (b). The target value was changed under the PforR from 47 percent to 30 percent (due to the unexpectedly low 8 percent baseline value); however, it was not formally revised under the Project. (c). This indicator is measured using the average of two indicators from the PIBS2: (i) percent of beneficiaries who know one program objective (71.7) and (i) percent of beneficiaries who know the correct value of payment (85.9). The average value was 78.8 percent, which exceeds the target. Page 18 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Justification of Overall Efficacy Rating 40. The overall efficacy is rated Substantial. Tables 3 and 4 outline and map the project attribution. Further details are included in Annex 1: Results Framework. C. EFFICIENCY Assessment of Efficiency and Rating 41. Efficiency is rated Substantial. This ICR analysis of the Project’s efficiency draws on the Economic and Financial Analyses (EFA) for the original Project and the AF. The two EFAs relied largely on global accumulated experience with CTs around the world and their success in reducing poverty and boosting human capital development. The AF further noted results from the CT-OVC impact evaluation which had demonstrated impacts on poverty reduction, increased quality of household consumption, and spillover effects such as increased activity in local economies (see details in annex 4). 42. Because of the broadened focus of the GoK’s SP initiatives, the consolidation of the CT programs under the NSNP, the parallel support from the World Bank-financed PforR, and limitations in the Project’s original EFA, it was not possible to report on some traditional measures of efficiency. The efficiency analysis therefore examines the Project’s cost effectiveness through briefly reevaluating the global experience with CTs; examining results generated by the Project’s investments; and assesses aspects of design and implementation that contributed to efficient or inefficient use of resources. 43. While the original project specifically targeted objectives attributable to the CT-OVC program, the IDA AF (US$10 million) helped harmonize and unify the NSNP as a whole and facilitated additional outcomes/results which exceeded the additional resources allocated to the Project. Specifically, the AF improved efficiency of the NSNP over the course of implementation by fostering the use of cost effective, streamlined and innovative measures, and strengthening the capacity of institutions and systems to utilize these measures. Interventions financed by the AF further contributed to meeting high-quality results related to the DLIs under the PforR operation. 44. Under the AF, the Project financed an expansion of TA to strengthen GoK’s coordination, management, and delivery of the NSNP, of which the CT-OVC program is a part. With the expanded scope, project benefits exceeded beyond the CT-OVC program and original project expectations to the GoK’s harmonized portfolio of CT programs. With this in mind, the efficiency analysis examines the Project’s costs and investments, in relation to the benefits and results generated for the CT-OVC program and the NSNP as a whole. 45. As noted above, under the AF, the original Project scope was significantly expanded and delivered outcomes and benefits beyond the costs and service fees associated with the restructurings, thereby having a positive impact on the Project’s efficiency (paragraph 13 of Annex 4). The AF strengthened the capacity of institutions and systems and has contributed to meeting high quality results related to the DLIs under the PforR. Under Component 1, the AF has been instrumental to foster efficiency gains by setting up a systems approach for the NSNP, adopting technology-based processes, and improving beneficiary selection. This includes the creation of an SR for all NSNP programs, development of an integrated MIS for the three programs under the MLSP and harmonized targeting across all NSNP Page 19 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) programs, contributing to substantial savings in the identification and enrolment of beneficiaries. The GoK also established a G&CM system to enable beneficiaries and others to submit grievances or updates on household data. Along with the introduction of two-factor electronic payments, these initiatives ensured that CTs became more secure, accessible, and predictable. 46. The economic rationale, as determined in the original EFA, remained strong. There are an estimated 2 million OVCs in Kenya, out of 18.6 million children ages between 0 and 14. OVCs therefore, constitute 10.8 percent of all children or 4.4 of the total population. Poverty rates for OVCs are slightly, but significantly, above the average child, with the poverty headcount rate for OVCs at 47 percent compared to 41 percent for all children ages 0-14 years. In addition, poor OVCs are in general poorer than the overall population of poor children with 12.5 percent living in extreme poverty, compared to 10.2 percent of overall children. 9 47. In relation to the estimated caseload, the CT-OVC Program currently covers 340,000 households – approximately 17 percent of all OVCs in Kenya. About 81.4 percent of OVC beneficiary households, surveyed for PIBS2, are within the harmonized PMT range indicating that beneficiaries are poor and deserving of support. 10 The surveys have also shown that the beneficiary benefit ratio (the proportion of the target population receiving transfer payments) increased from 70 percent before the Project in 2007/08 to 94 percent in 2016/17. The Project, therefore, demonstrates allocative efficiency in relation to its reach to the targeted population. 48. In terms of estimated Program impacts, the Project’s inputs converted into tangible and significant results. Under the original Project, a portion of IDA funds covered CTs of KSh 1,500 for 60,000 households. IDA support generated additional DP’s resources for CTs and allowed greater fiscal space for Program financing from the GoK, helping finance around 150,000 households up to FY2012/13. Under the AF, co-financing from DfID (US$56.4 million) supported 40,000 additional households with stipends of KSh 2,000 per household and leveraged enhanced GoK financing to the NSNP overall, with a cumulative number of 340,000 OVC households supported by Project closure. The Project’s investments in systems and institutional strengthening also enabled the GoK to increase coverage and financing of the CT-OVC and the NSNP overall. From 2007/08 (before the Project) until 2017/18 (the last year of the DIFD-financed TF), funding for the CT-OVC transfers increased from KSh 249 million to KSh 8,654 million. Of this, the GoK's contribution increased from KSh 150 million to KSh 7,694 million (representing an almost staggering 5,000 percent increase). 49. Increased school enrolment and completion of basic education enhanced coverage translating into measurable results with limited additional costs, since penalties for non-compliance with the conditions related to school enrollment were not implemented across the program. Based on PIBS2, almost 93 percent of beneficiaries reported positive impact on school attendance and performance for school-going children. Equally, the Kenya Poverty and PGA 2015/16 found that, overall, the NSNP programs had a significant and positive impact on child enrolment. More positive concrete impacts of the Program, range of welfare indicators and perceptions of the CT-OVC Program impact are discussed in 9World Bank estimates, using KIHBS 2015/16 data. 10PIBS2, October 2018. Estimated derived using the proposed new PMT formula; the computation of the percentage of beneficiaries who are within the harmonized PMT range used the Scoring Coefficients for the Harmonized PMT. The cut-off for the harmonized PMT yet to be decided. Page 20 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) more detail in paragraphs 8-9 of Annex 4. These indicators of reduced poverty and inequality substantiate the argument that the Project’s investments, together with the support of the GoK and DPs, has generated positive social and economic results commensurate with the Project cost/inputs. These results are expected to further yield numerous generational dividends and highlight the Project’s efficiency in relation to the conversion of IDA investments into substantial and tangible benefits. 50. It is also clear that the Project contributed significant economic benefits and value for money in relation to program implementation and administration. Administrative costs for running the Program were steadily reduced from 30 percent in FY07–08 (before the Project) to 2 percent during the last year of the Project 11. This decline is consistent with the potential cost savings generated by the Project’s investments in systems and processes. However, while administrative costs generally declined as safety net programs scaled up, the program’s allocation toward administrative expenditures need a substantial boost for the GoK to adequately implement the Program. In Africa, for example, administrative costs represent an average 17 percent of program spending. 12 D. JUSTIFICATION OF OVERALL OUTCOME RATING 51. The overall outcome rating is Satisfactory for the following reasons: • The relevance of the project objectives is rated High as it was and is well aligned with the country and World Bank strategies (paragraph 35). • Efficacy is rated Substantial because the majority of the targets set out in the PDO-level indicators both under the original Project and the AF as well as the intermediate indicators were achieved (paragraphs 36-40). • Efficiency is rated Substantial given that: (a) the number of beneficiaries and counties supported well exceeded original targets; (b) the strengthened technical skills under Component 1 improved efficiency through both improved processes and procedures across all the NSNP CT programs, through a strategy of consolidation and harmonization among the various programs; and (c) the Project contributed significant economic benefits and value for money as the administration costs for running the Program were steadily reduced from 30 percent in FY07–08 (before the Project) to 2 percent during the last year of the Project (see also paragraphs 42-51, 63 , 73 and Annex 4). E. OTHER OUTCOMES AND IMPACTS (IF ANY) Gender 52. Although the Project and the AF were not gender tagged, the Project benefited a large number of women and generated positive gender impacts. According to project data, about 870,000 of project beneficiaries are women (calculated by dividing the total number of beneficiaries—1.7 million—by two) 11World Bank estimates based on information provided by the SAU during the ICR mission. 12Beegle, Kathleen, Aline Coudouel, and Emma Monsalve. 2018 “Realizing the Full Potential of Social Safety Nets in Africa”. Africa Development Forum. Washington, DC: World Bank. Page 21 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) although they are not directly targeted as main beneficiaries in any targeting criteria. It should be noted that other sources mention that two-thirds of caregivers are women, indicating that a large proportion of the OVC Programs’ 340,000 beneficiary households are headed by women. The latter number seems appropriate given the role of women in society and their strong child-caring responsibilities. Implementation support visits to the field confirmed that overall, women are actively involved in the NSNP, both as household caregivers and as members of Beneficiary Welfare Committees (BWCs). This has been reported to enable women to increase control of household budgets and participation in income- generating activities. Evaluations have found that the CT programs reduced the likelihood of young women becoming pregnant and improved mental health outcomes, such as lower rates of depression, greater belief in self-agency and self-efficacy, and more positive views of the future. The HSNP impact evaluation further found positive impacts on purchasing power of women and their visibility as economic actors, and improved standing and status in their households. 13 53. Operational monitoring, 14 however, identified certain groups of women as potentially being excluded across the CT programs. These groups included single mothers in North Eastern Kenya, sex workers, who are regarded as social outcasts and therefore considered not deserving of assistance. OVCs whose parents have separated were provided as an example of not being eligible: “In instances where women from poor backgrounds have separated from their husbands or are single parents on other grounds, their children’s vulnerability is heightened. However, because both parents are still alive, they are not seen to meet the criteria for CT-OVC, because program officers place more emphasis on orphans than on vulnerable children”. 15 Although the CT-OVC Program has a well-defined criterion spelled out in the OM, such findings should continue to be monitored as the Program moves forward to inform future program adjustments as relevant. Institutional Strengthening 54. The technical assessment carried out for the PforR found that while the five CT programs 16 that constituted the NSNP were generally well designed, the CT-OVC delivered more efficient and effective support to households because of the level of resources dedicated to building programs systems (see paragraph 48, ensuring adequate staffing levels, improving M&E, introducing clear targeting mechanism, and so on. The AF contributed to ensuring that all the CT programs benefited from similar levels of support and facilitated harmonization among all NSNP programs. 55. The Project has been instrumental in strengthening the GoK’s capacity in the various institutions to formulate and implement policies. It has supported systems’ strengthening, by among others, biometric enrolment and targeting mechanisms. It also supported the preparation of framework to combat error, fraud, and corruption, as well as a comprehensive FM action plan. Strengthened FM system has introduced improved controls and reduced overall fiduciary risk of the NSNP. It has supported the development and implementation of and a consolidation strategy as well as the development of an 13 Oxford Policy Management. 2016. Impact Evaluation of the Kenya Hunger Safety Program Phase 2. 14 PricewaterhouseCoopers, Operational Monitoring Report 4 for the National Safety Net Program, 2017. 15 PricewaterhouseCoopers, 2017. 16 At the time of the assessment there was a fifth CT program – the Urban Food Subsidy Cash Transfer Program. It was however discontinued and the NSNP currently includes only four CT programs. Page 22 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) integrated MIS. Further, it supported a strategy of refining and harmonizing processes and procedures across all NSNP CT programs. 56. The Project brought about significant institutional strengthening through processes changes, as discussed in other parts of this report. The Project also provided extensive training in various aspects of the Program, including M&E, MIS, procurement, FM, and awareness sessions for beneficiaries on, among others, immunization, nutrition, and HIV/AIDs prevention. Poverty Reduction and Shared Prosperity 57. According to the Kenya Poverty and Gender Assessment (KPGA) 2018 17, Kenya saw a steady reduction in the poverty rate between 2005/06 and 2015/16, but progress is Modest. Over that period and consistent with the overall robust economic growth observed, the country has been able to reduce the national poverty headcount rate by more than 10 percentage points from 46.8 percent in 2005/06 to 36.1 percent in 2015/16. Despite this reduction in the incidence of poverty, the absolute number of poor people declined only marginally, from 16.6 million in 2005/06 to 16.4 million 10 years later. A look at the absolute number of the poor reveals that the number of people living below the poverty line increased in urban and NEDI counties, from 2.3 million to 3.8 million and from 2.4 million to 3.2 million, respectively, whereas it decreased in rural and non-NEDI counties. 58. The CT-OVC Project reached the poorest and most vulnerable households in 47 counties, greatly exceeding the number of households and counties planned for at appraisal. About 81.4 percent of OVC beneficiary households surveyed for the PIBS2 are within the harmonized Proxy Means Test (PMT) range, indicating that beneficiaries are poor and deserving of support. The PIBS1 and PIBS2 found strong positive impacts of the NSNP on overall well-being of beneficiaries, including consumption, household dietary diversity, school attendance and performance, and household health. III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 59. Project preparation was undertaken smoothly with no major issues. GoK’s commitment was evident from its level of involvement during the design/preparation stage. The original and revised objectives were clear and achievable. The design was simple and straightforward because it built upon an existing program with clear operational procedures spelled out in a comprehensive OM. An MIS had already been put in place for targeting, enrolment, complaints mechanisms, and updates on beneficiary households. Further, an in-depth review of the CT-OVC pilot had been conducted and the recommended adjustments were to be supported under the Project. The targeting mechanism had been evaluated and found to be functioning reasonably well while improvements were to be supported under the Project. Finally, an efficient payment system was necessary to provide regular payments to beneficiary households. While it was found that the arrangements with the Postal Corporation of Kenya (PCK) had been successful, other payments mechanisms were also to be tested under the Project. The Results Framework for the Project was well-designed and included indicators that were aligned with the 17 World Bank Group. 2018. Kenya Poverty and Gender Assessment 2015/16, A Decade of Progress and the Challenges Ahead. Report No: AUS0000466. Page 23 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) development objective. However, some of the indicators did not have real baseline values but were estimated using the most recent KIHBS data (2005/6), which caused some challenges during implementation when the real baseline values in some cases were higher or lower than expected and targets therefore had to be amended. The Project was prepared in close collaboration with a number of DPs actively supporting OVCs. UNICEF, which had been active from the pre-pilot phase, collaborated closely in the design, as did DFID, and SIDA. In the pre-pilot phase, support from DPs was already reasonably well coordinated. 60. Although program implementation had been successful at the time of project preparation, it was correctly acknowledged that the scale-up would be complex and demanding. The PAD included a table with identified risks and mitigation measures. Risks related to: (a) challenging technical and design issues (Substantial); (b) inadequate governance environment, including fiduciary risks (High); (c) limited implementation capacity and poor sustainability (Substantial); and (d) difficulties with DP coordination/harmonization (Substantial). The overall risk rating was correctly noted as High. B. KEY FACTORS DURING IMPLEMENTATION 61. The Project was designed as a CT operation. Nonetheless, the Project included some intermediate targets which were also relevant to a conditional cash transfer (CCT) type operations. An agreement was reached with the GoK during implementation to test the introduction of penalties to determine their effect on the impact of the Program. Because there was no conclusive research on the impact of penalties on improvement of key health and education indicators, it was agreed that the impact evaluation for the Project would aim to measure such impact. The pilot was implemented for about 18 months (2015/16, but faced challenges related to data collection, accurate implementation of penalties, and timely payments. Additional TA was engaged to support implementation, but after 18 months a review indicated that the pilot should start a fresh or be discontinued. A decision was made to discontinue it and focus resources on other aspects of the Program as the GoK had more pressing priorities and did not consider penalties as one of them. 62. It was important to place the implementation of the Project in the context of the broader NSNP agenda and make adjustments to the Project accordingly. The NSNP was established to realize key objectives of the National Social Protection Policy (NSPP) with regards to social assistance. In 2013, the main objective of the NSNP was (and continues to be) to improve the well-being of and increase resilience among specific vulnerable groups to reduce poverty and vulnerability. It was therefore important to make adjustments to the Project, including the PDO to better reflect this broader development in the SP sector and ensure relevant support to the broader NSNP. DP support to the NSNP included the: (a) World Bank- financed PforR; (b) World Bank and DFID financing for the CT-OVC Project; (c) Swedish support to other CT-OVC programs (provided through UNICEF); (d) direct support from UNICEF to the CT-OVC Program; and (e) DFID and Australian Aid financing for the HSNP. 63. The Project provided capacity building for implementation across the NSNP and implementation delays could negatively affect the NSNP as a whole. Any delays in the procurement of TA under the Project had a material impact on the wider NSNP and, at times, were a cause of slow progress toward achieving the DLIs under the PforR. At times, the GoK also faced implementation challenges operating the four programs, but additional support was brought on board, including employment of 210 Page 24 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) temporary officers to further support NSNP implementation. Additionally, external consultants were brought on board and hands-on support by the World Bank was provided. Support for capacity building under the AF has gone a long way to ensure that activities under the NSNP were carried out effectively despite delays in the CTs discussed elsewhere in this document. Delayed and unpredictable transfers created challenges for beneficiaries and reduced potential program impact. The main implementation challenges were, however, reduced significantly once the SAU was established and greater harmonization of the NSNP implementation and operational procedures was realized. The SAU was responsible for overall implementation and the technical support provided under the Project strengthened capacities directly to the SAU and thereby the Ministry. The performance of the SAU was overall impressive. High level of ownership and commitment ensured smooth implementation of a complex operation with various sources of funds within the Project itself and other projects financed by the World Bank/DFID and other DPs. 64. Implementation was closely monitored by the World Bank together with other DPs. Since preparation of the PforR, design and implementation support from the World Bank has been carried out for both operations and reported together in Aide Memoires. This ‘joint approach’ of reporting was positive because it ensured harmonized dialogue and implementation of the two operations. However, at times, the Aide Memoires did not sufficiently distinguish between the two operations, but this was corrected during implementation and the Implementation Status and Results Report (ISR) reporting is considered adequate, highlighting key issues throughout implementation. 65. As discussed in other parts of this report, one issue that prevailed throughout implementation was the challenge to deliver timeliness of payments to beneficiaries. Both the GoK and DPs recognized that timeliness of payments was critical for the success of the Project and the NSNP. It was acknowledged that Ministry of Labor and East Africa Community (Ministry implementing the CT-OVC Project at the time) and MDP (Ministry implementing the HSNP at the time) needed to work more closely with the National Treasury to achieve timely payments. Key challenges were identified, and a number of agreements were reached to address the issue. Payment procedures were streamlined, the SAU was created (preparing one consolidated payroll for all programs), changes were made in the budget planning and requisition of funds from the National Treasury, and in 2017 the SAU finally met the performance standards for timely payment to beneficiaries for three consecutive payment cycles for the first time since the inception of the NSNP. More recently, PIBS2 reported that 80 percent of beneficiaries surveyed reported that payments had arrived on time. The timely payments were also a result of more bank agents being made available closer to households, thereby reducing the time and costs of transport. 66. The GoK also made efforts to address low awareness among beneficiaries and establish a sound G&CM. Based on PIBS1, it became evident that awareness among beneficiaries of program objectives was really low. The Project therefore supported the development and implementation of the Inua Jamii’s BOS. The BOS intends to guide how to effectively reach out and communicate with NSNP beneficiaries, especially at the county and sub-county levels. The Project also supported the establishment of a G&CM system at the national level and commencement of decentralization of the same to enable smoother resolution and reporting of cases at the county level. The G&CM system helps the GoK better respond to beneficiaries with enhanced transparency and accountability. Beneficiaries may pursue various avenues: (a) a toll-free helpline; (b) talk to a member of BWC; (c) talk to a Chief or Assistant Chief; (d) contact Page 25 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) subcounty officers (SCO); (e) contact a county coordinator (CC); (f) use a drop box to submit a complaint, update, or feedback (available in CC and subcounty offices); (g) send an e-mail; or (h) write to the SAU. IV. WORLD BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 67. The theory of change for the Project is clear and straightforward and adequate and measurable indicators were identified to monitor progress toward achieving the PDO and to enable adequate M&E throughout implementation. As discussed earlier, changes had to be made to some of the indicators during implementation due to indicators being too difficult to measure (such as the targeting efficiency) or due to indicators combining several aspects into one (for example, the indicator measuring coverage and timeliness in one and which was later split into two). Effective M&E arrangements were put in place to regularly collect data and to enable easy analysis to inform program management and necessary changes throughout implementation. The combination of the GoK’s own internal monitoring system and the external monitoring arrangements, including regular operational monitoring reports, an impact evaluation, and the PIBS, are also considered adequate. As part of the AF and the restructuring in 2013, the Results Framework was also adequately amended to reflect and align with the GoK’s aims for the NSNP (see paragraph 28). The overall M&E design was comprehensive and strongly focused on generating impact results, which would inform decision making and guide program modification as needed. Over US$5 million or 25 percent of the costs of Component 1 were allocated to M&E activities under the original project and the AF. The Central Program Unit and, later, the SAU were tasked with the overall M&E coordination/implementation role. In retrospect, the Results Framework should have been revised when some indicators were changed for the PforR since it created inconsistency in one of the targets. M&E Implementation 68. The GoK was faced with a number of challenges related to M&E implementation but brought on board TA to support the resolution and key surveys reported important results. The M&E implementation was monitored systematically through continuous dialogue with GoK counterparts and both the GoK and World Bank brought on board TA to resolve challenges, mainly related to the impact evaluation and the PIBS. For both surveys, the procurement processes to bring on board the service providers were delayed. The baseline for the impact evaluation was not completed until 2015 and the impact evaluation end line and the second PIBS survey were barely finalized before project closure. Despite the challenges, the impact evaluation and PIBS reported important results. 18 The PIBS provided valuable data on beneficiary satisfaction and other relevant indicators with an overall high satisfaction level from beneficiaries. The engagement of an external firm to undertake the PIBS enabled provision of independent and rigorous data for monitoring performance on the implementation of the NSNP and the beneficiaries’ satisfaction with the services provided by the CT programs, including OVC. Results from the PIBS2 are discussed in detail under Efficiency (paragraph 49-50). The external monitoring reports and operational monitoring also provided important data to report success on meeting program objectives, 18Two rounds of surveys were undertaken for the PIBS: the baseline survey (PIBS1), carried out in 2015/16, and the follow-up survey (PIBS2), carried out in 2017/2018. Page 26 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) identify key challenges and inform necessary program adjustments. The GoK has regularly engaged external firms for operational monitoring on program implementation over the duration of the Project. During the past two years, in particular, regular external operational monitoring of the NSNP has taken place. Each report has focused on pre-identified key areas of interest including program harmonization, payments, G&CM systems, targeting, communication, and MIS, among other topics. These reports have been reviewed carefully by the GoK and have provided support for changes in program implementation addressing identified challenges. M&E Utilization 69. Monitoring and progress reporting of each indicator was used to inform project management and decision making and was included in Aide Memoires and ISRs. M&E under the Project for example informed decisions on design and DLIs included under the PforR. The fact that both projects were closely monitored ensured that results under this Project played a key role in the implementation of the PforR. Monitoring facilitated discussions with the GoK regarding outcomes. Since Component 1 supported the implementation of the PforR, monitoring of compliance with each target was closely followed and reported. The SAU was successful in developing an integrated MIS for the NSNP. These efforts should not be underestimated. A great deal of efforts was displayed to make changes to the systems to ensure that various information systems that had been in operation for the different CT programs were put under one single umbrella. In the past, the various systems “did not talk to each other”, therefore, assessment and results of performance of the various programs could not be fully compared. Justification of Overall Rating of Quality of M&E 70. The overall rating for M&E is Substantial and the M&E design both for the original project and the AF was sound. Implementation had some shortcomings as there were problems with data quality and the analysis conducted by the two firms engaged to implement the impact evaluation surveys. Nonetheless, with the help of the World Bank, the SAU moved quickly to engage additional TA to help improve the quality of the analysis. Utilization is rated Substantial, as the data collected were used to inform program management and decision making and influenced the choice of DLIs under the PforR. M&E ratings fluctuated between Moderately Satisfactory and Satisfactory. The main reason for the Moderately Satisfactory rating was the delay in the impact evaluation survey and the PIBS survey and the challenges faced with the contract management of the survey firms. The SAU, however, worked closely with additional TA and was able to conclude the two surveys before project closure. The M&E rating was upgraded to Satisfactory in the last ISR of December 2018. Follow-up Operation 71. Following the overall success of the CT-OVC, the GoK prepared the KSEIP with support from the World Bank. The KSEIP’s PDO is to strengthen delivery systems for enhanced access to social and economic inclusion services and shock-responsive safety nets for poor and vulnerable households. It supports the GoK to build upon the achievements made under the NSNP and CT-OVC through putting in place credible delivery systems for SP services. It supports results in three areas: (a) enhancing institutional capacity and further strengthening of SP delivery systems, particularly the coverage and functionality of the existing SR as well as to continue improving the efficiency of the payment and Page 27 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) grievance management system; (b) investing in the scale-up of existing nutrition-sensitive safety net and testing of customized economic inclusion models as a complement to the regular CTs to improve human capital and self-sufficiency of the poor and vulnerable households; and (c) improving shock responsiveness of the safety net system by expanding its coverage as well as strengthening financing arrangements for timely support to the affected households to improve their resilience and coping with recurrent climate- induced droughts. It is an IDA-financed investment project with DLIs in an amount of US$250 million. DFID is also co-financing the Project with a grant of GBP 66 million through a World Bank-managed TF. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 72. The Environmental Assessment rated the original project as Category ‘C’ as it was expected to have no environmental effects. The Social Assessment for the original project did not trigger any safeguards, but the AF credit approved in 2013 did. The Social Assessment of the original project concluded that a well-designed CT program could effectively reach the poorest households, redistribute income for the poor and thus raise their consumption levels. It also noted the strong evidence from countries worldwide showing that CTs could improve school enrolment and attendance rates and similarly may contribute to positive health outcomes. The AF triggered OP 4.10 (Indigenous Peoples) as the Environmental and Social Systems Assessment carried out for the PforR operation found that that CT programs sometimes create tension within communities because of the small number of households that benefit in the context of widespread need. It also identified the need to strengthen the participation and voice of poor and vulnerable groups in the NSNP. Thus, a Vulnerable and Marginalized Group (VMG) Framework was prepared by the GoK in accordance with the requirements of OP 4.10. 73. A Social Safeguards Review was carried out in July 2016 for the NSNP, including the CT-OVC Program. The review found no evidence that minority groups were being systematically excluded. Also, the revised consolidated OM included the core principle of non-exclusion and equity. Some dissemination workshops were carried out for the OM, at the time. Awareness raising was not carried out specifically for VMGs (although in 2018 awareness creation was carried out as part of BOS implementation). Round 4 of the operational monitoring looked at VMG inclusion and concluded that there was no systematic exclusion of VMGs, but more attention was needed to create awareness on barriers and ways to overcome them. VMGs may be excluded due to lack of awareness, or not being regarded as ‘deserving’ by their communities. The third and fifth operational monitoring reported no broad negative social impacts on the program, for example, conflict, negative attitudes/violence against beneficiaries, tensions with families, disincentives to productive work, or traditional support systems. The reports identified the need to strengthen communications and feedback on the complaints mechanism and to carry out more awareness in some counties on eligibility criteria for CTs. As the OVC Program will continue beyond the closing date, it is hoped that the remaining recommendations will be implemented, including the following: (a) the one- pager will be widely distributed to the rest of the VMG counties, (b) awareness sessions for COs on VMG inclusion will be carried out during BOS workshops, and (c) safeguards counterparts are identified in the State Department for Social Protection (SDSP) to ensure that all departments are aware of and follow up social inclusion issues. The rating for safeguards fluctuated between Moderately Satisfactory and Satisfactory. It was upgraded to Satisfactory in the December 2018 ISR. Page 28 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Procurement 74. Procurement under the Project was carried out in accordance with World Bank Guidelines and Procurement Plans were updates at least annually. All contracts other than those procured on the basis of International Competitive Bidding and consulting services followed the procedures under the GoK’s Public Procurement and Disposal Act of 2005. Procurement was initially the responsibility of the DCS and later the SAU, because the SAU took over the management of the NSNP, including the CT-OVC Project. The SAU managed procurement well. Procurement was rated Satisfactory during most of the project life with a short period during 2010/11 when it was rated Moderately Satisfactory. There was, however, a problem with the selection of firms to design and implement the surveys, as noted elsewhere in this report. Aide Memoires in 2017 and 2018 noted that there were no major concerns with procurement. Delays in procurement often related to insufficient funds included in the GoK’s budget or supplementary budgets. By late September 2018, several procurement packages for vehicles were still pending, because the National Treasury had not yet provided the SDSP with the authority to incur expenditures. Once the approval was given, an extension of the closing date of the TF to December 31, 2018, was processed to enable the SAU to complete the procurement of these packages. Nevertheless, due to the short time left until closing date, some vehicles could not be delivered on time and a balance of US$ 76,644 has been refunded to the World Bank. Financial Management 75. FM of the Project used country FM systems. The use of country FM systems is a positive achievement under the Project as the Project worked and strengthened country systems rather than use a parallel system that in many cases could be more “efficient”, but it leaves a vacuum once the financing of a project is concluded. IDA and DFID funds were disbursed under existing country systems. Funds were factored into the national budget and flowed through the GoK’s National Treasury and Exchequer systems. Planning and budgeting followed the GoK procedures, including the budget calendar. The CT- OVC Program resources allocated by the GoK and DPs provided the framework for the annual work plans and budgets. Budgeting implementation was monitored monthly through the GoK’s Integrated Financial Management Information System, bimonthly beneficiary payment cycle reports and reconciliations, and through quarterly interim financial reports. The overall budget was made up of capital transfers to households and administrative costs, including program operational costs, commission to PSPs, at the national, county, and sub-county levels and TA. In terms of CT, the budget was based on the estimated number of beneficiaries to be covered or enrolled in a particular year and associated operational costs per county and sub-county. The budget was harmonized and compiled by the SAU for inclusion in the ministry’s annual budget by March 31 every year. 76. Several reviews were conducted on the CT-OVC MIS and payroll. These identified anomalies, including irregularities in the beneficiary enrolment and payment system, culminated in payment of beneficiaries without names and ID numbers. According to the World Bank’s FM implementation support and supervision report of September 2018, these weaknesses were addressed through the development and implementation of an action plan developed jointly by the SAU and the World Bank. Fiduciary systems are also being strengthened as part of a specific DLI focusing on the implementation of critical FM actions under the AF of the PforR. In addition, the new PSP solution requires all beneficiaries to have bank accounts. During the rollout process, beneficiaries are required to provide proof of life and a significant Page 29 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) number of households are being removed from the payroll since the beneficiary is no longer alive. The new solution will also remove the need for extensive reconciliation at the end of each payment cycle, since funds are delivered to an account and will not be refunded to the ministry (which was previously the case if a beneficiary did not collect their payment). A more robust FM Policies and Procedures Manual was also developed as part of a DLI, introducing more comprehensive internal control procedures. 77. The Office of the Auditor General audited the project financial statements for the year ended June 30, 2017. The audit expressed a qualified opinion on the accounts for both the IDA Credit and the Grant. The qualification related to, among others, unsupported county expenditures and lack of explanation of some expenditures. These audit issues were however subsequently addressed. The audit for the year ended June 30, 2018, was also finalized. The audit expressed a qualified opinion on the financial statements and the Special Account Statement. The basis of qualified audit opinion included, among others: (a) non-financial assets procured during the year and lack of a complete fixed asset register on asset management; (b) funds transferred from the closed IDA counterpart project account and some assets procured in its lifetime not disclosed; (c) failure to provide CT reconciliation reports by the PSPs; (d) continuation of payments for some deceased beneficiaries; (e) delays by PSPs to submit reports on transfers to beneficiaries; and (f) overpayment of bank commission to Kenya Commercial Bank. Based on these and other audit findings an FM action plan was agreed upon with the World Bank and 80 percent of the priority FM actions have been implemented to date with the remaining 20 percent expected to be achieved by end of June 2019. In September 2017, the GoK took over financing of the 40,000 CT-OVC households which had been financed by DFID. This was a major achievement, which meant that 100 percent of the CTs for the CT-OVC Project were financed by the GoK. It demonstrates the GoK’s commitment and ensures sustainability of the Project. The GOK’s takeover of funding of CTs resulted in DFID needing to cancel the remaining amount of funds for CTs. It was agreed that instead of restructuring the Project at the time, the balance would be canceled at the time of Project closure. Subsequently around US$4.3 million was cancelled. C. WORLD BANK PERFORMANCE Quality at Entry 78. This evaluation rates the quality at entry as Satisfactory. The project design was consistent with the GoK’s strategy and the objectives of the Country Assistance Strategy 2004–08, which emphasized investing in people and governance. The World Bank team worked closely with the GoK to design a project addressing long-standing and new challenges facing the country since the post-election violence of 2008. The strategy for the Project was based on analytical and operational work carried out by the World Bank, which addressed poverty issues and inequality. It was recognized by the GoK, DPs, and the World Bank team that the World Bank could add value to the program due to its operational expertise on CTs in Latin America, Asia, and elsewhere in the world. The team ensured a participatory approach during project preparation at different levels of government, as well as with DPs. Quality of Supervision 79. The World Bank’s performance in ensuring quality of supervision is rated Satisfactory. A field- based World Bank team, supported by consultants, provided regular monitoring. The team provided timely and effective advice to counterparts. The task team leader (TTL) was changed four times because Page 30 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) of staff movement. All four TTLs were based in the field and the final TTL-shift was to a staff who had been involved with project preparation and supervision, which ensured continuity. TTLs maintained regular communication with the GoK and ensured that issues that arose were dealt with promptly and systematically. The project team responded quickly to the country’s request for AF to provide continuation and expansion of the TA necessary to support implementation of DLIs under the PforR. Supervision was carried out jointly with the PforR, ensuring consistency of approach and overall economies of scale. During implementation, the World Bank had 20 implementation support missions and several shorter technical missions. The team also maintained a productive working relationship with DPs and conducted joint supervision missions. The quality of the Aide Memoires was good, and the skill mix in the team was appropriate. The ICR mission held meetings with all the DPs involved in the Project, as well as being active in other CT activities. DPs expressed their appreciation of the World Bank team. They recognized the participatory approach of the World Bank which ensured that all were informed of developments and included in decision making when appropriate. 80. Procurement and FM were monitored regularly throughout the life of the Project. Both FM and procurement issues were identified by the World Bank team on time. The team provided assistance to the counterparts on ways to resolve issues or by encouraging the hiring of additional procurement staff/consultants to ensure that procurement issues were handled on time. In the case of FM, and on the basis of a number of audit observations, which highlighted fiduciary risks of the programs, TA was engaged to support the GoK in the development of a time-bound FM action plan (Aide Memoire of June 2017). The FM actions formed part of the evidence for one of the DLIs and focuses on overall system strengthening for long-term sustainability of strengthened fiduciary controls. Over 80 percent of the agreed FM actions had been achieved at the time of this writing (with 100 percent expected by June 2019 - ahead of the set target of December 2019 for NSNP). The World Bank team has monitored compliance with the steps and provided guidance when required. Justification of Overall Rating of World Bank Performance 81. Overall rating of the World Bank’s performance is Satisfactory for the above-mentioned reasons. D. RISK TO DEVELOPMENT OUTCOME 82. Risk to the development outcome is rated Negligible. The GoK has shown its commitment to maintaining and expanding the NSNP. This has been demonstrated in two ways. First, the commitment to expand the program and ensure consistency among new SP interventions. For example, the PforR built on achievements under the Project and in parallel, the AF supported the implementation of DLIs under the PForR. Similarly, the KSEIP expands approaches under the Project to ensure better returns. Second, GoK’s increased budget allocated to the various programs. Earlier difficulties with delayed CTs to beneficiaries have improved considerably and the new payment mechanism has been rolled out and enables timely payments of the CTs for the final three payment cycles of 2018/2019. However, there is still a critical need for the National Treasury to ensure timely transfer of funds to SDSP to enable timeliness of payments once the new payment solution is up and running. The November 2017 Aide Memoire reported that the GoK has successfully taken over the financing of (a) 40,000 CT-OVC households previously financed by DFID (disbursement-linked result [DLR] 1c) and (b) 11,000 households under the Page 31 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) HSNP previously financed by DFID (DLR 1e), demonstrating the GoK’s strong commitment to sustain the Program. 83. As recognized in the PAD for the KSEIP, the GoK has made significant progress in strengthening its social safety net system and establishing sound procedures through the NSNP. The NSNP seeks to improve the welfare of and increase resilience among specific groups to reduce poverty and vulnerability in Kenya by creating a framework around which the four main CT programs can be better coordinated and harmonized. 84. Further, as discussed in annex 4, there also continues to be strong DP commitment to SP in Kenya, which has translated into financial commitments from the GoK. Coverage of the NSNP has increased exponentially. The NSNP now reaches around 1.2 million households (almost 5 million people) from around 240,00 households in 2013. And the GoK finances almost 90 percent of program beneficiaries (compared to 38 percent in FY13). The GoK has further committed to take over 74 percent of the HSNP CT costs (currently at 64 percent) by FY20, such that the GoK would finance about 96 percent (about K Sh 25 billion) of the total NSNP cost. Under the KSEIP, this will be further increased to the GoK taking over 100 percent of the total NSNP cash payments to beneficiaries by July 2019. V. LESSONS AND RECOMMENDATIONS 85. The following are the key lessons learned. 86. A good design at preparation that takes into account lessons from other similar operations inside and outside the country is important. However, continuous investments in upgrading and improving delivery systems is equally critical to ensure effective implementation. For example, the original project design envisioned support to pilot an alternative payment mechanism to the PCK. At the time of project closure, the GoK was rolling out the third payment solution, which has advanced to a competitive model offering beneficiaries a choice in service provider between four different banks and providing them with a bank account. In 2019, 77 percent of beneficiaries reported having received payments on time compared to 51.5 percent in 2016 19 and the newly introduced two-factor authentication system has improved service to the beneficiaries, ensuring that CTs reached the intended beneficiary and the distance they had to travel to collect their payment reduced. The distance traveled by beneficiaries to the nearest pay point reduced with 61 percent of beneficiaries reporting that they traveled more than 6 km in 2016 compared to 15 percent in 2019. 87. Involving community representatives in the targeting processes for CTs increases ownership and community buy-in. Involving community representatives at different levels has increased community ownership and understanding of the targeting process, although more needs to be done to increase such understanding among the communities. For example, the HSNP experienced complaints and concerns raised by the communities related to the targeting process. The National Drought Management Authority therefore placed more emphasis on its communication strategy and involved community representatives in the targeting process, which reduced the level of distrust and complaints from communities. The 19 PIBS1 (2016) and PIBS2 (2019). Page 32 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) ongoing rollout of the BOS aims to increase community and beneficiary awareness, including around key program objectives and targeting. 88. Adequate targeting and recertification are essential to ensure maximum impact. Audit observations and external and internal monitoring have highlighted concerns about ineligible households enrolled in the NSNP. For example, there were reports about households continuing in the program despite their OVC turning 18 years, households who were reported to not actually have an OVC, or a severely disabled person or household remaining in the OPCT despite the older person having deceased. In any CT program, there is a certain degree of inclusion and exclusion errors and the status of the households keep changing, which is why there is a need to regularly recertify beneficiary households. The GoK has made commitments to address these issues. The rollout of the new payment mechanism requires proof of life and IDs for registration and this has demonstrated that some households should have been exited due to the older person passing away. The SDSP is also in the process of updating the Harmonized Targeting Tool, using the new KIHBS data (2015/16) and is finalizing a recertification strategy with rollout planned to commence next fiscal year. 20 In addition, under the KSEIP, the GoK is committed to rolling out a social registry which will be used for identification of beneficiaries for all future targeted programs and facilitating regular updates. 89. Strong government ownership and leadership is key to successful implementation. An important result under the PforR has been for the GoK to advance the agenda on harmonizing the CT programs. Progress in this area was only possible once there was strong ownership and commitment from the leadership in the SDSP taking on the harmonization agenda and making effective changes. Among others, the creation of the SAU, with the core mandate to manage the NSNP (as opposed to DCS which had several other mandates) has played a critical role in the success of the NSNP. Specifically, the provision of dedicated staff for the NSNP has enabled higher capacity as officers can focus on one specific area and undertake their tasks efficiently. Fully deployed payment officers enable swifter preparation of payrolls and separate MIS officers enable disaggregation of duties minimizing fiduciary risks. Specific officers fully deployed to key functions of MIS, G&CM, M&E and targeting has enabled important progress on results achieved. Lastly, strongly engaged and dedicated senior leadership including the Cabinet Secretary and Principal Secretary raises the visibility and profile of the program and increases ownership at the highest level and likelihood of continued future financing and commitment from the National Treasury. 90. Finding a streamlined and systematic solution to ensure timely payments to beneficiaries is critical. In 2016, an analysis of the steps required to complete a payment cycle found that 24 internal steps were required with a total of 90 days, making it practically impossible to make timely payments every two months. These steps were streamlined, reducing the time to 54 days. Subsequently in May 2017 for the first time since the establishment of the NSNP, the GoK managed to meet performance standard for timely payments for three consecutive payment cycles. However, payments became a challenge again when National Treasury did not prioritize disbursements to the SDSP (although challenges during the last year were more related to operational issues and not lack of priority from National Treasury). Therefore, there is a need to find a more sustainable solution and the National Treasury has agreed to include all CTs 20 This is also a DLI under the PforR. Page 33 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) on the recurrent budget instead of development budget, thus increasing the likelihood of timely payments because there are development financing budget cuts in times of budgetary constraints. 91. Improving or introducing new delivery systems may temporarily disrupt program implementation but has important benefits in the long run. The introduction of the new two-factor authentication payment mechanism in late 2015, disrupted program implementation because the SDSP faced several challenges with the two PSPs selected to undertake payments. Various challenges with collecting the biometric data were encountered (including fingerprints that could not be read or biometric machines malfunctioning). Once the new mechanism was rolled out, however, the logistical challenges decreased, and officers could focus more on other aspects of implementation, including internal M&E and G&CM cases. Once again in late 2018, when the new PSP solution was rolled out, implementation was disrupted, because all officers had to fully focus on migrating the over 800,000 households into the new mechanism across four different banks. Introducing competition across banks and enabling beneficiaries a choice of provider is however expected to significantly improve services. Evidence from field visits already indicates that beneficiaries have been able to choose a provider who is much closer in proximity than previous providers, thereby reducing the cost and time to access the pay point. 92. Strong institutional setup, including MIS and fiduciary systems is essential to improve efficiency and reduce risks related to implementation. Since the establishment of the SAU, implementation of the Project improved significantly. For example, 80 percent of the agreed priority FM actions have been implemented and therefore, the SAU is now doing regular reconciliations on time, something which had not happened since inception of the new payment mechanism in 2015. The new consolidated MIS is able to generate one single payroll compared to four different payrolls previously. This significantly reduced the time spent on payroll generation and reduced the risk of errors in the process. 93. Co-responsibilities with penalties should only be introduced if there is sufficient government ownership and commitment and sufficient human and budget resources to monitor compliance and implement supply-side interventions. Penalties were introduced twice into the program to test its impact, but it did not work because the DCS (and later the SAU) were not convinced that it would work and therefore did not fully commit themselves to follow up with county and subcounty officers to monitor compliance and enforcing penalties. Delays in payments also further challenged implementation, and there was insufficient capacity on the ground to undertake regular compliance checks with health clinics and schools. 94. Close collaboration with DPs played a critical role to the success of the Project. Working together with DPs not only contributed greatly to additional resources for the program but ensured that all collaborated to strengthening the GoK’s SP system as a whole and particularly the NSNP. Apart from monetary contributions, collaboration with DPs ensured a joint approach to supporting the GoK SP system and joint support from DPs also enabled timely response to requests from the GoK for necessary TA for project implementation. Any TA provided was aligned with support from other DPs. For example, UNICEF participated actively on project design and preparation. DFID was fully involved in supporting implementation and participating on joint implementation-support missions. WFP participated in discussions to ensure that their potential financial support was aligned with the Project and the PforR. Page 34 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) 95. Designing interventions that complement one another generate higher benefits. Since the PforR operation was designed in close collaboration with the AF to the Project, resources under the Project could be used to ensure high-quality results achieved under the PforR, which would otherwise have been challenging because it was a pure PforR with no Investment Project Financing resources. Specifically, the AF under the Project provided the technical assistance and overall institutional strengthening to support the implementation of DLIs under the PForR. For the Project, it was also beneficial to be aligned with the PforR because higher outcomes than originally expected could be achieved (including higher coverage of beneficiaries and important results in system development as mentioned elsewhere in this report). The new IDA-financed KSEIP includes interventions that expand on benefits generated by the Project and enables the SDSP to continue to ensure further progress in results areas related to the PforR and the Project. Moreover, having one World Bank team responsible for SP support to the GoK ensures consistency of approach and ongoing communications with the GoK as well as alignment of objectives and results. Page 35 of 66 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: To increase social safety net access for extremely poor OVC households Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Additional number of OVC Text 30000 50000 365000 350,000 beneficiary households receiving cash transfers 31-Mar-2009 31-Dec-2013 30-Dec-2018 31-Dec-2018 Comments (achievements against targets): The original target of the Project was 50,000 households, which was overachieved by far, given that the program currently has around 350,000 households registered in the MIS. The end target of the Project was however revised as part of the restructuring of the Project in December 2016 to 365,000. This target was therefore not fully achieved, mainly because some households have been exited from the program and no resources have been allocated for further expansion. The indicator is therefore considered mostly achieved, although it is worth highlighting that the program reaches more than six times the original target. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Proportion of OVC in Text 78 90 95 95 beneficiary households aged 6-17 years who have 27-Feb-2009 31-Dec-2013 31-Dec-2018 30-Nov-2018 completed or are enrolled in basic education Page 36 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Comments (achievements against targets): The 78% baseline refers to an estimate from the KIHBS data. The actual baseline value from the 2012 baseline impact evaluation is 94%. The draft end-line survey data confirms that the achieved value is 95% (and ongoing TA is in the process of reconfirming this value). At this stage the indicator is therefore considered achieved, but will be revised if TA confirms otherwise. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Project beneficiaries Number 0.00 478950.00 1825000.00 1750000.00 08-Oct-2009 30-Dec-2016 31-Dec-2018 30-Nov-2018 of which female Number 0.00 239475.00 912500.00 875000.00 08-Oct-2009 30-Dec-2016 31-Dec-2018 30-Nov-2018 Comments (achievements against targets): This number refers to the total number of CT-OVC beneficiaries (calculated as 350,000 households multiplied by 5). The number of female beneficiaries is calculated by dividing total beneficiaries by 2. Objective/Outcome: To build the capacity of the government to more effectively deliver the National Safety Net Program. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent of all NSNP payments Percentage 26.00 70.00 70.00 13.00 disbursed to payment service providers on time 30-Oct-2013 31-Dec-2018 31-Dec-2018 30-Nov-2018 Comments (achievements against targets): PIBSII confirms that 80% of beneficiaries reported that they received payments on time. The program has however experienced delay in payments during the last year and only 13% were made on time during the last three payment cycles before project closure. The delays are however still much less than earlier in program implementation and the main reason for the current Page 37 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) delay is the ongoing rollout of the new payment mechanism. Once the mechanism is fully rolled out it is expected that the payments will be timely and beneficiaries will enjoy a higher quality service since they have a choice in payment provider and are able to switch provider once a year. The indicator is however not considered achieved since the majority of the payments during the Project were not made on time and the end target was not achieved. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent of NSNP Beneficiaries Text 30 50 85 82.00 who conform to the program targeting criteria 31-Mar-2009 31-Dec-2013 31-Dec-2018 31-Oct-2018 Comments (achievements against targets): The PIBS I survey reported this number to be 60.8%. The PIBS II survey reported this number as 82.4%, but it should be noted that there were several challenges with measuring this value. This is one of the reasons it was decided, as part of the AF for the PforR in April 2017, that this indicator in the NSNP PforR framework would be replaced by an indicator related to recertification (due to the technical challenges with accurately measuring targeting conformity through PIBS). Further the PIBS II survey does not seem to provide comparable information to the baseline. SAU has prepared a recertification strategy which is ready for sign off and roll out of the same is expected to commence second half of 2019. The indicator is therefore considered mostly achieved. A.2 Intermediate Results Indicators Component: Component 1: Policy Development and Institutional Strengthening Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Social Protection strategy Text Social Protection Framework fully Social Protection New SP Policy and policy framework strategy in draft form developed Strategy adopted framework and SP established Strategy in process. Page 38 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) 31-Mar-2009 30-Dec-2011 31-Dec-2018 30-Nov-2018 Comments (achievements against targets): A Social Protection policy framework was adopted in June 2011. The Social Protection Secretariat is currently updating the policy framework and is preparing a Social Protection strategy. A workshop on the draft SP policy was undertaken September 5-7, 2018. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent of NSNP progress Text N/A 100 100 100 reports produced by the MIS, according to agreed 31-Mar-2009 31-Dec-2013 31-Dec-2018 30-Jun-2018 reporting standards Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent of NSNP Percentage 40.00 75.00 90.00 100.00 beneficiaries for whom payments are made 30-Jun-2013 30-Dec-2016 31-Dec-2018 30-Apr-2018 electronically through payment services providers using two factor authentication Comments (achievements against targets): The DLI has been achieved, but is still tracked. In the past three payment cycles, 100% of payments have been made using two-factor authentication. Page 39 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent of OVC payments Percentage 26.00 60.00 60.00 0.00 disbursed to payment service providers on time 30-Jun-2013 30-Dec-2016 31-Dec-2018 30-Nov-2018 Comments (achievements against targets): For the last three payment cycles prior to Project closure 0% of the CT-OVC payments were made on time. Component: Component 2: Program Implementation in Selected sub-counties Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of NSNP Text 37 60 60 89.35 beneficiaries who know program objectives and 28-Feb-2013 30-Dec-2016 31-Dec-2018 30-Nov-2018 entitlements Comments (achievements against targets): There are two different indicators in PIBSII used to measure this indicator; (i) knowledge of payment value (86%); and (ii) knowledge of at least 1 objective for the program in which a household is enrolled (92.7%). A simple average is therefore 89.35% and the indicator is considered overachieved. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Proportion of OVC living in Text 19 50 60 67 Page 40 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) beneficiary households 0-17 27-Feb-2009 31-Dec-2013 31-Dec-2018 30-Nov-2018 years holding birth certificates Comments (achievements against targets): The baseline estimate comes from the CT-OVC pilot districts in 2009. The actual baseline value from the 2012 impact evaluation was however 40%. The end-line survey has confirmed a final achieved 67% and the indicator was therefore overachieved. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Proportion of OVC in Text 66 80 70 81.00 beneficiary households aged 0-5 years fully immunized 27-Feb-2009 31-Dec-2013 31-Dec-2018 30-Nov-2018 Comments (achievements against targets): The baseline estimate comes from the CT-OVC pilot districts in 2009. The actual baseline value from the 2012 impact evaluation was however 51.4%. The endline survey reported on an indicator of "vaccination card up to date" which applied to 81% of beneficiaries which is interpreted to be the same as fully immunized for a beneficiary's particular age. The indicator is therefore overachieved. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent of NSNP Text 15.00 47 47 30.40 beneficiaries who can name two means of making a 31-May-2013 30-Dec-2016 31-Dec-2018 30-Nov-2018 complaint Comments (achievements against targets): The PIBSII data shows that this indicator is 30.40% as an average for NSNP programs. This indicator is Page 41 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) therefore achieved as per the formally revised target for NSNP PforR (which was revised to 30% in 2018 due to the 8% baseline value). The target for the Project was however not revised since no restructuring was made that close to project closing. Page 42 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) B. KEY OUTPUTS BY COMPONENT Objective/Outcome 1: To increase social safety net access for extremely poor OVC households 1. Additional number of households receiving timely and predictable cash transfers Outcome Indicators 2. Proportion of OVC in beneficiary households aged 6–17 years who have completed or are enrolled in basic education 1.Proportion of OVCs living in beneficiary households 0–17 years of age holding birth certificates Intermediate Results Indicators 2. Proportion of OVC in beneficiary households age 0–5 years fully immunized 3. Percent of NSNP beneficiaries who know program objectives and entitlements 1.Scale-up of CT from 45,911 households in 25 districts in 2009 to 355,000 in 69 districts in 2018 Key Outputs by Component 2. Increase of budget for CT from K Sh 1.4 million in 2009 to K Sh 8.43 million in 2018 (linked to the achievement of the 3. Support through a number of activities toward scaling up of CT-OVC Objective/Outcome 1) 4. Funding of temporary staff to support enrolment at county and sub-county levels 5. Regular awareness sessions for beneficiaries on immunization, nutrition and HIV/AIDs prevention. Objective/Outcome 2: To build the capacity of the GoK to more effectively deliver the National Safety Net Program 1. Percent of all NSNP payments disbursed to payment service providers on time Outcome Indicators 2. Proportion of beneficiary households that meet targeting criteria 1. Social Protection Strategy and policy framework established 2. Percent of NSNP beneficiaries for whom payments are made electronically through payment services providers using two factor authentication. Intermediate Results Indicators 3. Percent of OVC payments disbursed to payment service providers on time 4. Percent of NSNP beneficiaries who can name two means of making a complaint 5. Percent of NSNP progress reports produced by the MIS, according to agreed reporting standards Key Outputs by Component 1. National Social Protection Policy developed with support of TA (linked to the achievement of the 2. Social Protection Secretariat was established Objective/Outcome 2) 3. Alternative payments mechanisms for beneficiaries were developed with support of TA Page 43 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) 4. A great number of training took place related to M&E, MIS, enrolment, targeting, Beneficiary Outreach Strategy geared to toward different officers at different level (nation and county/subcounty levels) 5. Procurement System strengthened with TA 6. Harmonized Operations Manual for all CT programs under the NSNP was developed with TA 7. Procurement of vehicles, motorcycles, computers, printers, laptops, and accessories 8. Development of Communications Strategy and promotional materials with support of TA 9. Training of officers on implementation of strategy 10. Design of website for the ministry. 11. Design and implementation of PIBSs 12.Upgrade of MIS 13. Upgrade of targeting mechanism 14. Development of FM Module linked to payment service providers 15. Provision of alternative internet connectivity at national level 16. Design and implementation of Impact Evaluations 17. Expansion of biometric enrolment programs to facilitate payment authentication and pilot alternative payment mechanisms 18. Refurbishment of offices to accommodate MIS 19. Improved targeting process for the NSNP. 20. Development of consolidation strategy for all CT programs Page 44 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) ANNEX 2. WORLD BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Supervision/ICR Emma S. Mistiaen Task Team Leader Joel Buku Munyori, Lucie Muchekehu, Dahir Elmi Procurement Specialists Warsame Henry Amena Amuguni Financial Management Specialist Banu Setlur Environmental Specialist Sarah Coll-Black Team Member Cornelia M. Tesliuc Team Member Dorothy Morrow Akikoli Team Member Nadia Selim Team Member Josphine Kabura Kamau Team Member Joyce Cheruto Bett Team Member Michael Mutemi Munavu Team Member Lilian Wambui Kahindo Social Specialist Kevin Sanya Heraniah Team Member Betty Hanan ICR Author Samia Ausaf ICR Team Member Nancy Gamusa ICR Team Member B. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY08 0 205.00 Page 45 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) FY09 55.828 247,330.20 Total 55.83 247,535.20 Supervision/ICR FY09 0 2,908.51 FY10 37.566 172,953.95 FY11 43.985 194,892.47 FY12 30.843 169,094.96 FY13 43.014 260,547.27 FY14 38.274 322,416.66 FY15 52.130 332,887.91 FY16 77.565 396,971.78 FY17 31.346 241,861.93 FY18 2.225 249,432.76 FY19 8.696 275,140.90 Total 365.64 2,619,109.10 Page 46 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) ANNEX 3. PROJECT COST BY COMPONENT Table 3.1. Project Costs by Component (IDA Financing, US$ millions) Actual Actual Percentage Original AF Percentage Component AF Original of Original Project (US$, of AF Project Project millions) 1. Policy 10.0 10.0 7.0 9.5 70 95 Development and Institutional Strengthening 2. Program 40.0 0.0 37.9 0.0 94.7 0 Implementation in Selected Sub- counties Total project costs 50.0 10.0 44.9 9.5 90 95 Table 3.2 Project Cost by Component (DFID Financing, US$ millions) Percentage of Component Original Project AF Actual Total DFID Financing 1. Policy Development and Institutional 2.5 8.4 14.6 134 Strengthening 2. Program Implementation in Selected 28.7 39.7 60.4 88 Sub-counties Total project costs 31.2 48.1 75.0 95 Table 3.3. Sources of Funding (US$ millions) Organization Original AF Actual GoK 30.0 381.5 IDA 50.0 10.0 54.4 DFiD 34.0 56.4 75.0 UNICEF 12.0 0.0 26.4 Total 96.0 66.4 155.8 Page 47 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) ANNEX 4. EFFICIENCY ANALYSIS Introduction 1. This ICR analysis of the project’s efficiency draws on the Economic and Financial Analyses (EFA) for the original Project and the AF. The two EFAs relied largely on global experience with CTs and their success in poverty reduction and boosting human capital development. The AF further cited results from the CT-OVC program impact evaluation, which had demonstrated impacts on poverty reduction, increased quality of household consumption, and spillover effects such as increased activity in local economies. 2. Under the AF, the Project financed an expansion of technical assistance to the GoK to strengthen its coordination, management, and delivery of the NSNP, of which the CT-OVC program was a part. With the expanded scope, project benefits exceeded beyond the CT-OVC program and original project expectations, to the GoK's harmonized portfolio of CT programs. With this in mind, the efficiency analysis examines the Project’s costs and investments, in relation to the benefits and results generated, for the CT-OVC program and the NSNP as a whole. 3. As per the ICR guidelines, the efficiency analysis should examine if the costs involved in achieving project objectives were reasonable in comparison with both the benefits and with recognized norms (“value for money”) and look at the extent to which the Project achieved the maximum possible benefits with the minimum possible costs. When applicable and practical, traditional measures of efficiency—such as net present value, economic rate of return or financial rate of return — should be used. However, because of the broadened focus of the GoK’s Social Protection initiatives, which the Project supported, including the consolidation of the CT programs under the NSNP, and limitations in the Project’s original EFA, it was not possible to report on some of these traditional measures of efficiency. The ICR efficiency analysis therefore examines the Project’s cost effectiveness through briefly reevaluating the global experience with CTs; examining results generated by the Project’s investments; and assessing aspects of design and implementation that contributed to efficient or inefficient use of resources. 4. While a parallel PforR operation was designed with financial support from the World Bank to incentivize the achievement of results related to the NSNP, critical investments were also required in implementation capacity to help ensure that the results under the PforR were achieved. This was one of the main justifications for processing the AF for the Project and despite the fact that it was not possible to disentangle the exact cost/inputs for each result and assess how it specifically led to a precise benefit/outcome, the efficiency analysis emphasizes critical results achieved with limited overall investments and concludes that the efficiency of the Project was substantial and that Project funds were used wisely and achieved good value for money. This is particularly based on the critical investments in the delivery systems compared to the significant results related to such investments focusing on: (i) strengthening program systems to ensure good governance; (ii) harmonizing CT programs to improve sector coherence; and (iii) capacity building for expanding CT programs to promote more comprehensive and equitable coverage. Background and Global Experience on CT Programs 5. Since the preparation of the original project EFA, social safety programs have been prolifically evaluated. Positive empirical evidence on the welfare effects of safety net programs has helped spur Page 48 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) substantial investment in these programs over the past decade. CTs have been estimated to reduce the incidence of absolute poverty (US$1.90 purchasing power parity per day) by 36 percent, and relative poverty (the bottom 20 percent) by 8 percent. 21 A meta-analysis conducted on impacts for seven African countries found that on average, household consumption increases by US$0.74 for each US$1 transferred. 22 The magnitude of this impact varies across countries, with the largest impact experienced by programs targeted at the poor. Most recent evidence confirms the positive and significant impact of CTs on increased school enrolment and attendance. Results on health, nutrition, and food security also continue to be positive and significant. Safety net programs also tend to increase the consumption of calories in targeted poorer households, resulting in frequent direct health impacts on young children. 23 Lastly, CTs have also proven to have major spillover effects on the local economy—for each US$1 transferred to a beneficiary household, the total income of the local economy may increase by more than US$1. 24 Economic Rationale for Investing in the OVC Program 6. The economic rationale, as determined in the original EFA, remained strong. There are an estimated 2 million OVCs in Kenya, out of 18.6 million children ages between 0 and 14. OVCs. Therefore, OVCs constitute 10.8 percent of all children, or 4.4 percent of the total population. Poverty rates for OVCs are slightly, but significantly, above the average child, with the poverty headcount rate for OVCs at 47 percent compared to 41 percent for all children ages 0–14 years. In addition, poor OVC are in general poorer than the overall population of poor children with 12.5 percent living in extreme poverty, compared to 10.2 percent of overall children. 25 7. In relation to the estimated caseload, the CT-OVC program currently covers 340,000 households —approximately 17 percent of all OVCs in Kenya. About 81.4 percent of OVC beneficiary households, surveyed for PIBS2, are within the harmonized PMT range indicating that beneficiaries are poor and deserving of support. 26 The surveys have also shown that the beneficiary benefit ratio (the proportion of the target populations receiving transfer payments) increased from 70 percent before the Project in 2007/08 to 94 percent in 2016/17. The Project, therefore, demonstrates allocative efficiency in relation to its reach to the targeted population. Estimated Program Impacts 8. The Project’s inputs converted into tangible and significant results. Under the original Project, a portion of IDA funds covered CTs of KSh 1,500 for 60,000 households. IDA support generated additional development partner resources for CTs and allowed greater fiscal space for program financing from the GoK, helping finance up to around 150,000 total households by FY 2012/13. Under the AF, co-financing 21 State of Social Safety Nets 2018, World Bank. 22 Ralston, Laura, Colin Andrews, and Allan Hsiao. 2017.” The Impacts of Safety Nets in Africa: What Are We Learning?” Policy Research Working Paper No. 8255, World Bank, Washington, DC. 23 State of Social Safety Nets 2015, World Bank. 24 State of Social Safety Nets 2018. 25 World Bank estimates, using KIHBS 2015/16 data. 26 PIBS2, October 2018. Estimate derived using the proposed new PMT formula; the computation of the percentage of beneficiaries who are within the harmonized PMT range used the Scoring Coefficients for the Harmonized PMT. The cut-off for the harmonized PMT yet to be decided. Page 49 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) from DFID (US$56.4 million) supported continued financing of the 40,000 households with stipends of KSh 2,000 per household and leveraged enhanced GoK financing to the NSNP overall, with a cumulative number of 340,000 OVC households supported by Project closure. The Project’s investments in systems and institutional strengthening also enabled the GoK to increase coverage and financing of the CT-OVC program and the NSNP overall. From 2007/08 (before the Project) until 2017/18 (the last year of the DFID- financed TF), funding for the CT-OVC CTs increased from KSh 249 million to KSh 8,654 million. Of this, the GoK’s contribution increased from KSh 150 million to KSh 7,694 million (representing an almost staggering 5,000 percent increase). 9. In line with the Project’s objective of increased school enrolment and completion of basic education, this enhanced coverage translated into measurable results with limited additional costs, since penalties for non-compliance with the conditions related to school enrollment were not implemented across the program (paragraph 94 of the main text). Based on PIBS2, almost 93 percent of beneficiaries reported positive impact on school attendance and performance for school-going children. The KPGA 2015/16 further found that, overall, the NSNP programs had a significant and positive impact on child enrolment. About 93.5 percent of NSNP beneficiary households containing school-age children had all those children enrolled in 2015/16. The program with the largest impact was the CT-OVC program; the impact of a household being an OVC beneficiary was an increase in the probability of all children being enrolled of about 3 percentage points. Further, the CT-OVC program had an almost 5 percentage point impact on the probability that no school-age child in the household was working. 10. In terms of perceptions of the CT-OVC program impact, the results were also positive. The PIBS2 acknowledged that the proportion of beneficiaries reporting positive impacts from the program on household welfare was very high for all the socioeconomic indicators measured: (a) 94.7 percent of beneficiaries reported positive impact on household consumption; (b) 92.6 percent of beneficiaries reported positive impact on household dietary diversity; (c) 96.5 percent of beneficiaries reported positive impact on school attendance and performance; and (d) 89.4 percent reported positive impacts on household health. 11. Using data from the KIHBS 2015/16, the KPGA further examines the impacts of the NSNP on a range of welfare indicators. Three main findings about the existing programs are that: (a) they are well targeted; (b) they have had positive effects on school enrolment, have reduced the probability of children working, and have reduced food insecurity; and (c) poor non-beneficiaries would benefit greatly from expanding such programs. The above indicators of reduced poverty and inequality substantiate the argument that the Project’s investments, together with the support of the GoK and DPs, have generated positive social and economic results commensurate with the Project cost/inputs. The above results are expected to further yield numerous generational dividends and highlight the Project’s efficiency in relation to the conversion of IDA investments into substantial and tangible benefits. Design and Implementation Changes Project Restructurings 12. As illustrated in Section B, the Project had three restructurings and they are all conisdered to have contributed positively to project efficiency. The Project’s first restructiring enabled the provision of additional funding to the DFID-financed Multi Donor Trust Fund (MDTF) and provided additional IDA funds Page 50 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) under the AF. The closing date of the orignal Project was extended by three years and the originial Project scope was significantly expanded. As seen in the section below (Additional Financing), this restructuring delivered outcomes and benefits beyond the costs and service fees associated with the restructuring, thereby having a positive impact on the Project’s efficiency. The second restructing adjusted the Results Framework to align with the PforR and improve measurability of key indicators; made necessary amendments to institutional arrangements based on changes in the MLSP; and extended the closing date by almost two years. This extension supported achievement of additional technical assistance which enabled achievement of high-quality results (which would otherwise not have been possible), related to among other things: (i) adoption of expansion plan and expansion of the NSNP coverage; (ii) achievement of timeliness of payments; (iii) adoption and implementation of the NSNP consolidation strategy and; (iv) strengthened fiduciary systems. By improving alignment with the on-going PforR, for which the Project intensively provided technical assistance, this restructuring further enhanced achievement of PforR DLIs and did not negatively impact project efficiency. On the contrary, it contributed to achievement of additional important results. Delays in the GoK’s approval to spend the remaining Recepient-Executed Trust Fund amount further prompted a third and final restructing to extend the closing date by two months. However, the extension enabled completion of all outstanding activities, including analysis of vital PIBS2 and impact evaluation data. The Project’s M&E rating was thereby upgraded to Satisfactory in the final (December 2018) ISR (from previous ratings of Moderately Satisfactory). The final extension therefore had no adverse impact on the Project’s efficiency. Additional Financing 13. While the original Project made specific investments targeted to the CT-OVC program, the IDA AF (US$10 million), representing a mere 16 percent increase in IDA resources, helped harmonize and unify the NSNP as a whole and facilitated additional critical benefits/outcomes which exceeded the additional costs/resources allocated to the Project. Specifically, the AF improved efficiency of the NSNP over the course of implementation by fostering the use of cost effective, streamlined and innovative measures, and strengthening the capacity of institutions and systems to utilize these measures. Interventions financed by the AF further contributed to meeting high-quality results related to the DLIs under the PforR operation. Below includes key support that the AF financed: • The AF made modest investments towards the creation of an NSNP consolidation strategy, related to technical assistance and the consultation process to finalize the strategy. The subsequent adoption and implementation of this strategy led to important results, demonstrating Project efficiency, including: the creation of the Social Assistance Unit (SAU); development of the consolidated operational manual for the three SAU programs; preparation of consolidated work plans, budgets, and performance targets; consolidated M&E reports; and development of an integrated program MIS. These measures introduced numerous efficiencies by cutting staff time and cost, and yielded benefits and savings that far exceed Project costs. • The creation of a consolidated program MIS for all SAU CT programs enabled merging of three separate SAU program payrolls 27 into a single payroll, generating efficiencies in payment processing. SAU 27 Payrolls for the CT-OVC, Older Person’s Cash Transfer and Cash Transfer for People with Severe Disabilities Page 51 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) now has one payment team, generating one single payroll from one consolidated MIS. This will continue to generate efficiency gains beyond the life of the Project. • The creation of a Single Registry for all NSNP programs, linked to the Integrated Population Registration Services (IPRS), only required modest inputs in terms of technical assistance for design, and maintenance and investments in equipment for the system. This has already and will continue to generate benefits in the long-run for the NSNP. It enables the GoK to determine who is in the NSNP programs and avoids duplication across programs. Since its establishment in November 2015, the SR has already generated benefits that outweigh the costs of investments in the system and will continue to do so through the more rigorous checks on enrollment and internal controls on the payroll. • The introduction of two-factor authentication electronic payment system required limited investments from the Project. Yet, it has ensured that CTs are more secure and accessible. Moving from a paper-based payment system (through the Post Offices) enabled stronger records for audit purposes and decreased the risk of error, fraud and corruption, since the new system more efficiently records all transactions electronically. The Project’s investments also generated an important improvement in timeliness. Under PIBS2, 77.3 percent of beneficiaries indicated that the last payment collected was received within the time when it was scheduled (an improvement from 51.5 percent in PIBS1). The electronic payment system also enabled payment providers to engage a number of agents to pay beneficiaries, thereby cutting the time and costs associated with the beneficiaries having to travel long distances and wait in long queues to receive the payments. 84 percent of beneficiaries had to travel less than 6 km to reach the nearest pay point per the PIBS2, compared to 39 percent in PIBS1. • The implementation of the Financial Management action plan strengthened fiduciary systems and generated savings from stronger controls and reduced leakages. This includes payroll reconciliation, approval of financial policies and procedures manual, and refund by one of the PSPs based on the reconciliation. Administrative costs of the implementing agency 14. The Project contributed significant economic benefits and value for money in relation to program implementation and administration. Administrative costs for running the program were steadily reduced from 30 percent in FY07/08 (before the Project) to 2 percent during the last year of the Project. 28 This decline is consistent with the potential cost savings generated by the Project’s investments in systems and processes. However, while administrative costs generally declined as NSNP programs were expanded, the program’s allocation toward administrative expenditures need a substantial boost for the GoK to adequately implement the program in future. In Africa, administrative costs represent an average 17 percent of program spending. 29 Project costs 15. The Project used allocated funds well. By the end of the Project all funds had been accounted for. 28World Bank estimates based on information provided by the SAU during the ICR mission. 29Beegle, Kathleen, Aline Coudouel, and Emma Monsalve. 2018. “Realizing the Full Potential of Social Safety Nets in Africa.” Africa Development Forum. Washington, DC: World Bank. Page 52 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Trust fund resources were also well utilized. Of the total US$ 139 million only around US$ 300,000 were refunded (some due to ineligible expenditures and some due to inability to spend before Project closure) and an amount of around US$ 4.3 million was cancelled under the Trust Fund. It is important to note, however, that this cancellation came about because of the important result achieved for GoK to take over financing of CTs for 40,000 CT-OVC beneficiary households, previously financed by DFID. Since this takeover was done earlier than originally expected, it generated a cost saving in terms of less funds used for CTs. The GoK was able to use some savings for additional technical assistance (through an overdraw from component 2) to, inter alia, facilitate achievement of results (including for the PforR’s DLIs), including implementation of critical Financial Management actions, contributing to strengthened fiduciary systems; and adoption of a Beneficiary Outreach Strategy, thereby increasing beneficiary awareness of the program. Page 53 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) ANNEX 5. BORROWER ICR Background Context 1. Kenya continues to experience steady economic growth and declining poverty incidence; however, inequality remains high across the country. Kenya is classified as a lower-middle-income country (gross national income per capita of US$1,146). Between 2004 and 2016, economic growth averaged 5.4 percent and is expected to reach 5.5 percent in 2018 and rise to 6.1 percent by 2020. Barring sharp oil price increases, inflation is expected to be within the GoK’s target of 5 percent in 2018. The National Treasury projected a decline in the fiscal deficit from 9.1 percent of gross domestic product (GDP) in FY2016/17 to 5.7 per cent of GDP in FY2018/19. Due to good economic performance, poverty incidence (measured against the official poverty line) is estimated to have dropped from 46 percent to 36 percent between 2006 and 2016, however with wide regional variation. While the Gini coefficient declined over the last 10 years, Kenya remains one of the most unequal countries in the world. 2. High rates of poverty and inequality exacerbate vulnerabilities, undermine investments in human capital, and limit the potential for the benefits of economic growth to be shared by all Kenyans. Many Kenyans continue to lack food security and good health care. The maternal mortality rate remains one of the highest in Africa at 488 deaths per 100,000 live births. Poor children, when compared to their better- off counterparts, are less likely to attend school, advance to the next school grade, be vaccinated, and have good health outcomes. Moreover, while poverty and vulnerability are highly correlated, over one- third of non-poor Kenyans are vulnerable because they are clustered just above the poverty line, meaning that even a moderate shock could push them below the poverty line. 3. Among human development challenges, high stunting and wasting rates in ASAL counties are of particular concern. While Kenya is on track to meet the World Health Assembly national targets for nutrition, there are serious shortfalls in some counties. For example, West Pokot and Kitui, have stunting rates nearly 20 percentage points higher than the national average. Moreover, the 2017 drought led to an increase in acutely malnourished children of 25 percent nationally and over 30 percent in three ASAL counties. 4. The number of OVC in Kenya is growing, an estimated 1.8 million children ages below 18 years are orphans having lost one or both parents. About two-thirds of the orphans in Kenya have lost their parents due to HIV/AIDS. Apart from the orphans, there are an increasing number of other ‘vulnerable children’ who are in need of care and protection. The National AIDS Control Council estimates that there are about 600,000 children who are vulnerable as their well-being and development are threatened for various reasons. 5. Families without one or both parents are poorer than others, especially in rural areas. Although traditionally community systems used to cater for such children, these mechanisms are being overstrained by the HIV/AIDS pandemic coupled with high levels of poverty. Therefore, there is increasing recognition that these children need special support to avoid development of ‘underclass’ and harness their human development capacity. Page 54 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Description of the Project 6. The CT-OVC project aimed to assist the GoK to support extremely poor and OVC gain access to basic services. This was to be done through strengthening the capacity of households taking care of OVC to be able to care and protect them within their families and promote their human capital by providing regular and predictable CTs. Since its inception, the Project was being implemented by the DCS until July 2016 when the SAU was created within the ministry and was mandated to coordinate implementation at the national level while the DCS officers continues with implementation at county and subcounty level. 7. The Project was to (a) support the GoK to expand the program coverage in terms of beneficiaries and area, (b) improve the targeting and enrolment mechanism, and (c) enhance the capacity of the GoK to manage and strengthen the program governance. 8. The original project for the CT-OVC IDA credit (IDA-45540) closed on December 31, 2013, and the AF under IDA Credit (IDA-53120) ended by December 31, 2016, while the World Bank Trust Fund (DFID) Grant (TF-097272) will close by October 31, 2018. Project Objectives 9. To provide an SP system through regular and predictable CTs to poor households living with and taking care of OVC. The aim was to strengthen the capacity of families to care for and protect OVC and promote human capital development. 10. Specific objectives of the Project were the following: • To increase the capacity of the DCS to manage the CT-OVC Program • To support the implementation of the program in the selected districts by providing timely and predictable CTs to selected households caring for OVC PROJECT COMPONENTS UNDER THE ORIGINAL PROJECT Component 1: Policy Development and Institutional Strengthening Subcomponent 1.1: Strengthening of the Capacity of the MGCSD to Coordinate SP interventions 11. The component was designed to support the development of a policy framework for coordinating existing SP programs. In addition, it was to support the overall planning and coordination of SP interventions in the country. Under this, an SPS was to be established within the ministry to play this role. Subcomponent 1.2: Strengthening the Capacity of the Department of Children’s Services to More Efficiently Manage the Program 12. This subcomponent was intended to strengthen the capacity of the DCS to effectively implement, coordinate, and monitor the implementation of the expanded program. Specifically, the support was to conduct trainings for implementing staff and creating and strengthening structures at the districts for effective program implementation and to support development of a Program Operations Manual and Page 55 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) training materials. In addition, the subcomponent was to support hiring of TA for procurement and FM. In addition, the subcomponent was to provide support for (a) strengthening links with complementary programs such as health and education for OVC and (b) coordinating with other human development promoting programs such as school bursary schemes. Subcomponent 1.3: Governance and Accountability 13. This subcomponent was to support a range of activities that were to further strengthen the governance of the program. These activities included development of a program communication strategy; an information, education, and communication (IEC) strategy; and promotional information materials. Further, the subcomponent was to support awareness sessions for beneficiary households to raise their awareness on their rights and responsibilities on OVC such as importance of immunization, nutrition, and other related knowledge to improve the life of OVC and household members. This awareness was to support a program oversight and accountability system by ensuring that beneficiaries and non- beneficiaries can make appeals and complaints. Subcomponent 1.4: Management Information, Monitoring and Evaluation 14. This subcomponent was designed to support strengthening of the program MIS to include new inbuilt targeting and enrolment infrastructure and an FM module that was linked with the PSP. The MIS also incorporated M&E modules for assessing the implementation of the program. Component 2: Program Implementation in Selected Districts Subcomponent 2.1 Cash Benefits for OVC Households in Selected Districts 15. This subcomponent was planned to directly support selected extremely poor households taking care of OVC for the four financial years, to build their capacity to care and protect OVC under their care. Subcomponent 2.2: Transaction Costs for Cash Transfer 16. This subcomponent was intended to cater for cost of delivering CTs to beneficiaries through contracted payment service providers and meeting other programs’ operational cost in selected districts. ADDITIONAL FINANCING 17. The AF to the CT-OVC Program was active from 2013 to 2016 and was meant to support provision of TA to the GoK to strengthen its coordination, management, and delivery of the broader NSNP of which CT-OVC was part. The PDO was to support the two components of the original project, with some revisions being done on Component 1. Therefore, the components under the AF are detailed in the following paragraphs. Page 56 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Component 1. Policy Development and Institutional Strengthening 18. This component was to support building the capacity of the GoK to implement the program by the subcomponents, as listed in the following paragraphs. Subcomponent 1.1: Strengthening Program Systems to Ensure Good Governance 19. The key activities under this subcomponent were (a) support for the expansion of biometric enrolment for the program beneficiaries to facilitate payment authentication and pilot alternative payment mechanisms, (b) procurement of equipment and vehicles for effective field operations, and (c) review of the targeting mechanisms and processes for CT-OVC and other NSNPs to improve and harmonize program targeting mechanisms and recertification of beneficiary’s eligibility. Subcomponent 1.2: Harmonizing Cash Transfer Programs to Improve Sector Coherence 20. This subcomponent was to support (a) development of a consolidation strategy for the programs under the ministry and (b) carrying out of program implementation and beneficiary satisfactory survey. In addition, the subcomponent supported review of the Program Operations Manual and developed one consolidated Operations Manual. Subcomponent 1.3: Providing Capacity Building for Expanding CT Programs to Promote More Comprehensive and Equitable Coverage 21. This subcomponent was to support rapid targeting of new program beneficiaries to be enrolled into the program and recruitment of temporary staff under the contract to support the department staff to implement the expanded program. Component 2. Program Implementation in Selected Sub-counties 22. This component entailed supporting the regular CT to poor and vulnerable households caring for OVC who were enrolled into the program through monthly stipends. PROJECT DESIGN, IMPLEMENTATION, AND OPERATIONAL EXPERIENCES 23. The PDO was to strengthen the capacity of households to care and protect OVC in the short run and build their human capacity in the long run. This objective was in line with Kenya Constitution 2010, Children Act, Social Pillar of Vision 2030, and the NSPP. The project design was appropriate as it enabled the GoK to support about 40 percent of OVC living in extremely poor households through CTs thus enabling them access to basic services. 24. The co-responsibility component of the Project was difficult to implement due to its design cumbersomeness and hence was discontinued after two pilots. Page 57 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) RESULTS FRAMEWORK AND OUTPUTS Table 5.1. Results Framework and Outputs under the Original Project Component Planned Outputs at Appraisal Actual Outputs Component 1: Subcomponent 1.1: Policy Strengthening of the Capacity Development of the MGCSD to Coordinate and Institutional Social Protection Interventions Strengthening 1.1.1: Technical Assistance to TA was engaged to assist the development of the SP Develop a Social Protection strategy and an NSPP that provided coordination Strategy and Policy Framework framework for SP interventions in the country. to Coordinate NSPPs for Coordination of SP Interventions 1.1.2: Support to the Social The SPS was established and given the mandate of Protection Sector Groups and coordinating SP interventions. However, sector groups Social Protection Secretariat in were not created since the SPS superseded it. the Ministry 1.1.3: Support for Testing TA was engaged to explore alternative payment Alternative Payment mechanisms with two-factor authentication for program Mechanisms for OVC and beneficiaries. Other Social Protection Programs Component 2: Subcomponent 1.2: Program Strengthening of the Capacity Implementation of the DCS to Manage the in Selected Program Districts 1.2.1: Build the Capacity of the A total of 109 officers were trained either locally or Officers Implementing the abroad on various aspects of the program, such as Program and National, designing and implementation of social transfer Provincial, and District Levels program, M&E, integrated information systems, and including the Locational project management, among others. The trainings Structures to Efficiently targeted officers in different cadres both at headquarters Implement the Program and field levels. Technical Assistance on FM TA on procurement was engaged to assist the GoK team and Procurement to address program procurement needs. TA on FM was engaged to assist in financial matters. TA to review and update the TA was engaged to assist the program in reviewing and Program Operations Manual updating the Program Operations Manual. The TA also and develop training materials developed training materials. Procurement of critical 55 computers, 10 photocopiers, 25 printers, 13 scanners, hardware and software and assorted accessories were procured. Procurement of motor vehicles 20 motor vehicles and 20 motorcycles were purchased to and motorcycles enhance field operations. Provide support for Program implementing officers conducted regular strengthening links with awareness sessions on beneficiaries particularly on complimentary programs such immunization, nutrition, and common infectious diseases health and education such as HIV/AIDS. Subcomponent 1.3: Page 58 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Component Planned Outputs at Appraisal Actual Outputs Governance and Accountability 1.3.1: Technical Assistance to TA was engaged to assist in developing the Program develop a Communication Communication Strategy and other program promotional Strategy, Design and materials such as IEC. Training was organized for the Development of Information, officers on the same. Education and Communication Materials. TA also assisted in creating a program website and link it to the ministry website. 1.3.3: Conducting A firm was engaged to conduct three rounds of program Beneficiaries’ Satisfaction implementation beneficiary satisfaction surveys (PIBS). Assessment Only two rounds were completed. Subcomponent 1.4: Management Information, Monitoring and Evaluation 1.4.1: TA to Support Upgrade TA was engaged to (a) upgrade the program MIS, (b) of the Program MIS to improve PMT score, (c) improve targeting mechanism, Incorporate Financial and (d) develop an FM module linked to the PSP. Management Modules Linked to PSP 1.4.2: Provide an Alternative A firm was engaged to provide alternative interment Internet Connectivity at connectivity at the national level. Some provincial and National, Provincial, and districts offices were connected through modems. District Level 1.4.3: TA to strengthen the A firm was engaged to conduct program operational and program M&E system for impact evaluation. efficient program implementation Subcomponent 2.1: Cash Benefits for OVC HHs in selected districts 2.1.1: Scale-up of program in The number of households was exceeded by the target terms of coverage by 56,000 year since 153,323 households were being supported by households in 25 selected 2013 in 69 districts. districts by 2013 Subcomponent 2.2 Transaction Costs for Cash Transfers 2.2.1: Funds for delivery of The funds supported the officers in the field to effectively cash transfers to beneficiary implement the program and also meet the cost of paying households commissions due to PSPs. Results Framework and Achievement under the AF Component Planned Outputs at Appraisal Actual Outputs Component 1: Subcomponent 1.1: Policy Development Strengthening Program Systems and Institutional to Ensure Good Governance Page 59 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Component Planned Outputs at Appraisal Actual Outputs Strengthening 1.1.1: TA to Support the A firm was engaged to conduct biometric Expansion of Biometric enrolment for beneficiaries who were paid Enrolment for the Programs to biometrically through the new payment Facilitate Payment mechanism. Authentication and Pilot Alternative Payment Mechanisms 1.1.2: Procurement of Additional 191 desktop computers, 20 photocopies, 263 Computers, Hardware, Software, printers, 350 laptops, 124 cameras, 1,545 and Related Equipment tablets, 38 scanners, 38 fireproof cabinets, and assorted accessories were procured. 1.1.3: Procurement of More 51 motor vehicles were procured. Motor Vehicles to Support Field Operations 1.1.4: Refurbishment of Offices Offices were refurbished to accommodate MIS to Accommodate MIS and data clerks in the National Social Security Fund - Block C 4th floor. 1.1.5: TA to Review Effectiveness A TA successfully reviewed the CT-OVC targeting of Targeting Process of CT-OVC process and documented proposal to improve and other NSNP Program to PMT. Improve Harmonizing Program Targeting Subcomponent 1.2: Harmonizing Cash Transfer Programs to Improve Sector Coherence 1.2.1: TA to Support in A firm was engaged to develop a harmonization Developing a Consolidation and consolidation strategy for the NSNP program Strategy for the Programs in the to improve targeting. Ministry 1.2.2: TA to Carry out PIBS for A firm was engaged to regularly carry out rounds the NSNP Services of PIBS to assess the levels of beneficiaries’ satisfactions on program services 1.2.3: Review and Update The Program Operations Manual was reviewed Programs Operations Manuals and consolidated into one manual for the Regularly consolidated cash transfer program 1.3: Capacity Building for Expanding the Cash Transfer Programs to Promote More Comprehensive and Equitable Coverage 1.3.1: Support Activities toward Over 220,000 new beneficiaries were enrolled Scale-up of program to Cover into the three programs. More Households 1.3.2: Funding for Temporary 210 temporary staff were recruited to support Staff to Support the Enrolment program implementation both at field and at Field Level national levels Component 2: 1.3.1: Strengthen the Ability of 40,000 households were supported through CTs Program Poor and Vulnerable Households Page 60 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Component Planned Outputs at Appraisal Actual Outputs Implementation in to Protect and Care for OVC and Selected Sub-Counties Promote their Human Capital 1.3.2: Support Administrative Funds were made available to (a) finance Activities Geared toward targeting and enrolment processes of new Delivery of Cash Transfer beneficiaries and (b) commissions to PSPs. PROJECT COORDINATION AND MANAGEMENT 25. From its inception, the Program was coordinated, managed, and implemented by the DCS staff at the national, provincial, and district levels up to 2013 when Kenya’s new constitution was promulgated, thus changing the country’s administrative units from province to counties and districts to sub-counties. Even then, the Program management remained with the DCS until July 2016 when the SAU was established and given the mandate of coordinating the three CT programs. Currently, implementation of the Program is coordinated by the SAU at the national level while the implementation at county and subcounty levels is still being implemented by DCS staff with liaison officers acting as a link between the SAU and the department. BORROWER’S OWN PERFORMANCE DURING PROJECT PREPARATION AND IMPLEMENTATION 26. The GoK has since the start of the Program remained committed to achieving project objectives as depicted by continuous and increased allocation of funds to the Program since 2005. The GoK has developed and continues to develop necessary policies to provide an enabling environment under which the program is implemented, specifically, the NSPP and other documents, which have facilitated program implementation such as the Program Operations Manual, consolidation strategy, BOS, among others. Currently, the GoK is in the process of amending the Social Assistance Act 2013 to be in line with new trends in the social assistance sector. The GoK has maintained its oversight role in monitoring implementation of the program. 27. The ministry implementing the program played its role of coordinating the program activities and ensuring their completion in time. Major accomplishments of the program so far are the following: • A robust MIS to process the program beneficiary’s data • Consolidation of the CTs into one program • Harmonization of the targeting methodology for all CT programs • Setting up of a G&CM mechanism • An elaborate M&E system for the program 28. The staff tasked with implementation of the program, although few in number, remain motivated to carry out the duties. The above aspects, coupled with the cooperation from different supportive sections of the ministry, made the successful implementation and achievement of its objectives possible. The GoK, through the implementing department, always strived to ensure that all that what was expected was done in time. Some of these were regular funds requests through preparation of quarterly IFRs and required internal and external audits. Page 61 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) 29. However, a co-responsibility aspect of the program, which was piloted midstream was found to be cumbersome to implement and was discontinued. This was because it relied on other agencies of the GoK for its implementation and was also quite costly because it required engagement of temporary staff at the community level to regularly collect data. PERFORMANCE OF WORLD BANK AND OTHER PARTNERS DURING PROJECT PREPARATION AND IMPLEMENTATION Assessment of World Bank 30. The World Bank Kenya country team worked very closely with the GoK right from the project design through to Implementation. The project design preparation was supported significantly by TA of the World Bank. The setting of the PDOs was done in consultation with the GoK to ensure they were in line with the GoK sector priorities of alleviating the situation of OVC in the country. The regular consultative meetings and periodic joint review missions held during the program implementation assisted in addressing various issues and bottlenecks experienced and suitable solutions found. 31. Changes made to the project subcomponents during the implementation were carefully made to ensure they maintained the original PDO. 32. The World Bank team took an active role in coordinating other partners, in particular UNICEF, DFID, and the WFP, whose technical and financial support greatly contributed toward improvement of the Program. Assessment of DFID 33. The DFID staff actively participated during the project design and implementation by offering their technical support in areas of need. Their periodic reviews and sharing of their findings greatly contributed to identifying program weakness, and recommendations made improved program delivery. DFID provided co-financing to the Project, which supported 40,000 beneficiary households through payments of monthly stipend up to June 2017 and further continued to strengthen the ministry capacity to implement the program through provision of necessary equipment and vehicles. 34. The relationship between the GoK and the World Bank together with other partners remained sound throughout the project period (2009-2018). The support received from the World Bank, DFID, and other partners has greatly supported expansion of the Program to be a national program. DESCRIPTION OF THE PROPOSED ARRANGEMENTS FOR ENSURING PROJECT SUSTAINABILITY 35. The GoK is very committed to sustain the achievement of the completed project and continue to expand the coverage of the program as evidenced by full takeover of financing of the program from the exchequer. In addition, the GoK has absorbed 210 officers who were previously on contract term into permanent term, which is a boost to the existing staff in the implementing departments. Further, the GoK is in the process of amending the Social Assistance Act 2013 to adequately anchor the funding into a law thus giving legal basis under which the program will continue to operate. The MLSP has also created the SAU and deployed staff at the national level with the sole mandate of coordinating implementation of the Page 62 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) program. The unit has developed a robust Program MIS, which automates all the Program processes right from household identification to final payment reconciliation and generation of reports. And a new payment model has been launched with multiple PSPs contracted to provide an account-based payment system that is more secure and accessible and that encourages beneficiaries’ savings. FINANCIAL EVALUATION OF THE PROJECT Total Project Cost to Date 36. IDA financing: • Original project was US$50 million. • DFID Trust Fund in the original project was US$34 million. • UNICEF funding was US$12 million. • DFID Trust Fund under the AF was US$56.4 million. • AF was US$10 million. • GoK contribution toward the Project was US$477million. Lessons Learned • Targeted CT programs are most appropriate for developing countries where poverty level is high and resources are limited. • CT programs can be used as a strategy to mitigate poverty among the poor communities. • Targeting process for CTs is transparently done by involving community representatives at different levels, which increases community ownership and reduces complaints. • CTs empower the poor and disadvantaged segments of the population. • CT programs promote micro-level investments through multiplier effects. • Regular and predictable CTs enhance dignity and credit worthiness among beneficiaries. Conclusion 37. The Project was designed according to the GoK priorities toward addressing poverty and vulnerability in the country. The objectives set were crucial, for they not only strengthened the capacity of the GoK to manage and coordinate the program but also supported the GoK’s rapid expansion of the program to cover all constituencies in the country. The Project played a central role in strengthening capacities of government institutions and systems to deliver cash transfer effectively and efficiently to the neediest households. Page 63 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) ANNEX 6. SUPPORTING DOCUMENTS 1. UNICEF Frequently Asked Questions - May 2006 2. World Bank Option Paper for CT-OVC - December 2006 3. Management Letters - 2013–18 4. Evolution of the GoK Cash Transfer Program - January 2007 5. DFID Fiduciary Assessment of the CT-OVC - April 2007 6. Integrated Safeguards Data Sheet - May 1, 2007 7. Operations Manual for Consolidated Cash Transfer Program (MLSP) - June 2017 8. Project Concept Note - April 7, 2008 9. Project Information Document (June 30, 2008) 10. Draft Communication Strategy for CT-OVC (MGCSD) - September 2008 11. Minutes of Quality Enhancement Review - October 2, 2008 12. Minutes of Project Concept Note - June 4, 2008 13. Project Appraisal Document (Report No. 44040-KE) - February 20, 2009 14. Financing Agreement - Credit 4553-KE – May 8, 2009 15. TF Administration Arrangement (TF 071488) - May 18, 2010 16. Additional Financing Decision Note - September 10, 2013 17. Project Paper for the Additional Financing (Report No. PAD731) - October 8, 2013 18. Decision Note on Additional Financing - September 10, 2013 19. Minutes of Negotiations for Additional Financing - October 2, 2013 20. Second Additional Contribution and Amendment to the TF071488 - November 1, 2013 21. Financing Agreement – Credit 5312-KE - December 5, 2013 22. Grant Agreement TF097272, December 19, 2013 23. Amendment to the TF 071477 Administration Arrangements - October 7, 2014 24. Annual Reports – NSNP - October 2016, 2017 25. Review of the Special Audit Report of the Auditor General - February 14, 2017 26. Special Audit Report: Refund of Questioned Expenditures - December 4, 2017 27. Operational Monitoring Report (Sixth Monitoring Cycle) - PricewaterhouseCoopers Limited – June 2018 28. Borrower’s Contribution to ICR - October 2018 29. Aide Memoires - 2006–18 30. Implementation Status and Results reports (ISRs) - 2009–18 31. Management and other important correspondence - 2013–18 32. Beneficiary Outreach Strategy and Action Plan - May 2018 33. Project Appraisal Document for the Kenya Social Protection and Economic Inclusion Project - October 2018 34. Program Implementation and Beneficiary Survey II - October 2018 Page 64 of 65 The World Bank Kenya Cash Transfer for Orphans and Vulnerable Children (P111545) Page 65 of 65